# import libraries

import pandas as pd
import numpy as np
import requests
import warnings
import matplotlib.pyplot as plt
import seaborn as sns
import re
from IPython.display import display, Image, IFrame, HTML
import json
import ast


from wordcloud import WordCloud
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
import string
import nltk
from collections import Counter
import plotly.express as px

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.cluster import KMeans
from sklearn.metrics.pairwise import cosine_similarity
import matplotlib.pyplot as plt
from sklearn.decomposition import PCA
from sklearn.metrics import silhouette_score

from sentence_transformers import SentenceTransformer
from sklearn.metrics.pairwise import cosine_distances
from sklearn.decomposition import PCA, KernelPCA
import hdbscan 
import umap
from tqdm import tqdm



# set options
pd.set_option('display.max_columns', None)
pd.set_option('display.max_colwidth',None)
#pd.set_option('display.max_rows', None)
import warnings 
warnings.filterwarnings('ignore') 

nltk.download('stopwords')
nltk.download('punkt')

[nltk_data] Downloading package stopwords to
[nltk_data]     /Users/bridgetnevel/nltk_data...
[nltk_data]   Package stopwords is already up-to-date!
[nltk_data] Downloading package punkt to
[nltk_data]     /Users/bridgetnevel/nltk_data...
[nltk_data]   Package punkt is already up-to-date!

True

#load to dataframe
#df = pd.read_csv('art_objects.csv')
df = pd.read_csv('HAM_7_5_25-2.csv')
dropped_cols = ['contextualtextcount', 'accesslevel', 'dateoflastpageview', 'totaluniquepageviews', 'markscount',
                'publicationcount', 'totaluniquepageviews', 'contact', 'colorcount', 'rank', 'id', 'state', 
                'verificationleveldescription', 'images','imagecount', 'worktypes', 'imagecount', 'classificationid',
                'standardreferencenumber', 'signed', 'relatedcount', 'verificationlevel', 'titlescount',
                'peoplecount', 'lastupdate', 'periodid', 'edition', 'lendingpermissionlevel', 'colors', 'groupcount',
                'dateoffirstpageview', 'imagepermissionlevel', 'mediacount', 'techniqueid', 
                'seeAlso', 'details', 'videos', 'audio']

df.drop(dropped_cols, axis = 1, inplace = True)
print(f"There are {df.shape[0]} objects in the collection")
print(f"The number of years that there was no data on accession year is {df['accessionyear'].isna().sum()}, which is {round(100* df['accessionyear'].isna().sum()/df.shape[0], 2)}% of the records")
df_acession = df[~df['accessionyear'].isna()]
print(f"The first year that Harvard aquired an object in the collection was {int(df_acession['accessionyear'].min())}.")
print(f"The number of years that there no is data on created year is {(df['dateend']== 0).sum()}, which is {round(100* (df['dateend']== 0).sum()/df.shape[0], 2)}% of the records. ")
print(f"The oldest object in the collection is from {int(df['dateend'].min())}.")

There are 246377 objects in the collection
The number of years that there was no data on accession year is 74150, which is 30.1% of the records
The first year that Harvard aquired an object in the collection was 1765.
The number of years that there no is data on created year is 70737, which is 28.71% of the records. 
The oldest object in the collection is from -6995.

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 246377 entries, 0 to 246376
Data columns (total 42 columns):
 #   Column           Non-Null Count   Dtype  
---  ------           --------------   -----  
 0   copyright        62070 non-null   object 
 1   creditline       246377 non-null  object 
 2   createdate       246377 non-null  object 
 3   division         238618 non-null  object 
 4   terms            237974 non-null  object 
 5   period           36438 non-null   object 
 6   totalpageviews   246377 non-null  int64  
 7   accessionyear    172227 non-null  float64
 8   classification   246377 non-null  object 
 9   primaryimageurl  222869 non-null  object 
 10  style            2251 non-null    object 
 11  commentary       2426 non-null    object 
 12  technique        168884 non-null  object 
 13  description      36690 non-null   object 
 14  medium           106110 non-null  object 
 15  title            246374 non-null  object 
 16  accessionmethod  246377 non-null  object 
 17  provenance       74945 non-null   object 
 18  related          84546 non-null   object 
 19  dated            231386 non-null  object 
 20  department       246377 non-null  object 
 21  dateend          246377 non-null  int64  
 22  titles           246374 non-null  object 
 23  people           214984 non-null  object 
 24  url              246377 non-null  object 
 25  century          226751 non-null  object 
 26  places           114042 non-null  object 
 27  objectnumber     246377 non-null  object 
 28  labeltext        1096 non-null    object 
 29  datebegin        246377 non-null  int64  
 30  culture          232561 non-null  object 
 31  exhibitioncount  246377 non-null  int64  
 32  objectid         246377 non-null  int64  
 33  dimensions       192318 non-null  object 
 34  marks            40956 non-null   object 
 35  exhibitions      19967 non-null   object 
 36  publications     23130 non-null   object 
 37  gallery          1698 non-null    object 
 38  groupings        65157 non-null   object 
 39  tags             303 non-null     object 
 40  contextualtext   1433 non-null    object 
 41  media            81 non-null      object 
dtypes: float64(1), int64(5), object(36)
memory usage: 78.9+ MB

df = df.sort_values(by='dateend')
df[:1]

print(f"There are {df[df.dateend == 0].shape[0]} objects that have a 0 value date")
print(f"There are {df[df.dateend.isna()].shape[0]} objects that have a NA value date")
print(f"There are {df[df.accessionyear.isna()].shape[0]} objects that have NA accession year")
print(f"There are {df[df.accessionyear== 0].shape[0]} objects that have a 0 value accession year")

There are 70737 objects that have a 0 value date
There are 0 objects that have a NA value date
There are 74150 objects that have NA accession year
There are 0 objects that have a 0 value accession year

year_df = df[df['dateend']!= 0]
plt.figure(figsize=(10, 6))
sns.histplot(data=year_df, x='dateend', kde=True, label = 'date created', bins = 1000)
sns.histplot(data=df_acession, x='accessionyear', kde=True, label = 'date acquired')
plt.title('Years When Objects Were Created And Acquired')
plt.xlabel('Year')
plt.ylabel('Frequency')
#plt.xlim(-2000, 2025)
plt.ylim(0, 5000)
plt.grid(True)
plt.legend(loc='upper left')
plt.show()

year_df = df[df['dateend']!= 0]
plt.figure(figsize=(10, 6))
sns.histplot(data=year_df, x='dateend', kde=True, label = 'date created', bins = 1000)
sns.histplot(data=df_acession, x='accessionyear', kde=True, label = 'date acquired')
plt.title('Years When Objects Were Created And Acquired')
plt.xlabel('Year')
plt.ylabel('Frequency')
plt.xlim(-1400, 1400)
plt.ylim(0, 1000)
plt.grid(True)
plt.legend(loc='upper left')
plt.show()

year_df = df[df['dateend']!= 0]
plt.figure(figsize=(10, 6))
sns.histplot(data=year_df, x='dateend', kde=True, label = 'date created', bins = 1000)
sns.histplot(data=df_acession, x='accessionyear', kde=True, label = 'date acquired')
plt.title('Years When Objects Were Created And Acquired')
plt.xlabel('Year')
plt.ylabel('Frequency')
plt.xlim(1400, 2025)
plt.ylim(0, 20000)
plt.grid(True)
plt.legend(loc='upper left')
plt.show()

print(f"{round(100* df['division'].isna().sum()/df.shape[0],2)} percent of records do not have a division.")
div_df = df[~df['division'].isna()]
div_df = pd.DataFrame(div_df['division'].value_counts())
div_df.plot.pie(y='count', labels= div_df.index, autopct='%1.1f%%', startangle=90, shadow=True, ylabel='')

plt.axis('equal')  # Equal aspect ratio ensures that pie is drawn as a circle.
plt.title('Object Division Distribution')

plt.legend(loc='center left', bbox_to_anchor=(1.5, 0.5))
plt.show()

print(f"There are {len(df.classification.unique())} different classes of objects")

fig, ax = plt.subplots(figsize=(10, 12)) 
class_df = pd.DataFrame(df['classification'].value_counts(dropna= False))
top_class_df = class_df.iloc[0:30]
ax = sns.barplot(x='count', y='classification', data=top_class_df, orient='h')
for container in ax.containers:
    ax.bar_label(container)
plt.title('Top 30 Object Classification Distribution')
ax.set_xlim(0,90000)
plt.show()

3.15 percent of records do not have a division.

There are 60 different classes of objects

plt.figure(figsize=(10, 60))

culture_df = pd.DataFrame(df['culture'].value_counts())
ax=sns.barplot(x='count', y='culture', data=culture_df, orient='h')
for container in ax.containers:
    ax.bar_label(container)
plt.title('Object Cultures')
ax.set_xlim(0,95000)
# Show the plot
plt.show()

def show_record(title):
    row = df[df['title']== title]
    print(row['title'].values[0])
    image_path = row['primaryimageurl'].values[0]
    display(Image(url = image_path, width = 400))
def show_label(title):
    row = df[df['title']== title]
    print(row['labeltext'].values[0])

show_record("Races, Negroes: United States. Virginia. Hampton. Hampton Normal and Industrial School: Agencies Promoting Assimilation of the Negro: Development of Social Standards Among the Negroes. Hampton Normal and Agricultural Institute, Hampton, Va.: Geography - Local Industry.")

Races, Negroes: United States. Virginia. Hampton. Hampton Normal and Industrial School: Agencies Promoting Assimilation of the Negro: Development of Social Standards Among the Negroes. Hampton Normal and Agricultural Institute, Hampton, Va.: Geography - Local Industry.

show_record('Negro Mounted Upon a Rearing Horse')
show_record("Negro in a Punt in a Landscape")
show_record("Negro Huts, near Wilminton, North Carolina")
show_record("Negro Page")
show_record("The Negro")
show_record('Lo, the Poor Indian')
show_record("The Franciscan and the Indian")
show_record("The Oriental Peasant and His Family")
show_record('Oriental Hunters Chasing Stags')
show_record('Oriental Hunters Killing Monkeys')
show_record("An Oriental Holding a Horse")
show_record("Oriental Dancer")
show_record("Mulatto and African Slaves")
show_record("Cripple with a Crutch and a Wooden Leg")

Negro Mounted Upon a Rearing Horse

Negro in a Punt in a Landscape

Negro Huts, near Wilminton, North Carolina

Negro Page

The Negro

Lo, the Poor Indian

The Franciscan and the Indian

The Oriental Peasant and His Family

Oriental Hunters Chasing Stags

Oriental Hunters Killing Monkeys

show_record('Two Women of Color, Dominica [French Mulatress of Dominica and a Negro Woman in their Proper Dress]')
show_record('Negro Soldier')
show_label('Negro Soldier')
print(" ")
show_record("Woman of Mixed Race")
show_label("Woman of Mixed Race")

Two Women of Color, Dominica [French Mulatress of Dominica and a Negro Woman in their Proper Dress]

Negro Soldier

Robert Smullyan Sloan was drafted into the army in 1943 and stationed in New York, where he illustrated educational materials for the armed services and produced posters for the U.S. Treasury’s war bond campaign. In this painstakingly detailed composition reminiscent of early Flemish painting, he portrays a U.S. Army private wearing a dress uniform with ribbons designating a Good Conduct Medal and participation in the European–African–Middle Eastern Campaign. The artist’s title, however, accords him only the designation of his race, using the language of that historical moment.
African Americans fought a two-front battle during World War II, facing enemies overseas and racism at home. The view out the window in Sloan’s work alludes to the distinguished soldier’s dreary prospects upon his return to the gritty urban environment of 125th Street in Harlem.
 
Woman of Mixed Race

A growing metropolis in the early 20th century, Berlin attracted diverse ethnic groups, particularly in the realms of art and theater. When he visited, Nolde avidly depicted scenes of modern urban life, including this portrait of a stage performer. The bright yellow circle surrounding the woman like a halo evokes a mirror, and the garish orange of the background reflects the gaslight backstage. Nolde’s title for this painting refers to the subject’s mixed-race heritage. In common use then, the term is now considered offensive for the way it devalues people based on their Black ancestry. The year Nolde painted this work, 1913, was the same year legislation was passed in Germany that made “German blood” the sole criterion for citizenship, as a way to exclude Germans of African descent — a law that was changed only in 2000. Represented here in vibrant colors, the unnamed subject lived and worked in Germany despite having no legal protection. Race is central to recent deconstructions of Nolde’s self-stylized myth. He openly supported the National Socialists, even after they officially denounced his work and included this painting in the infamous “Degenerate Art” exhibition in 1937. Ironically, in spite of the artist’s own antisemitism and his support for the Nazi party, his painting has become a familiar symbol of the art they so vigorously suppressed.

def search_string_in_dataframe(df, search_string, case_sensitive=False):
    """
    Searches for a string in any field of a Pandas DataFrame.

    Args:
        df (pd.DataFrame): The DataFrame to search within.
        search_string (str): The string to search for.
        case_sensitive (bool, optional): Whether the search is case-sensitive. Defaults to False.

    Returns:
         pd.DataFrame: A DataFrame containing rows where the string was found, or an empty DataFrame if not found.
    """
    mask = df['title'].astype(str).str.contains(search_string, case=case_sensitive, na=False).values
    #mask = np.column_stack([df[col].astype(str).str.contains(search_string, case=case_sensitive, na=False).values for col in df])
    
    return df[mask]

l = search_string_in_dataframe( df, "Seven little Indians" )
image_path =l['url'].values[0]
display(HTML(url = image_path))

i = search_string_in_dataframe( df, "Indian Chiefs Who" )
image_path =i['url'].values[0]
display(HTML(url = image_path))

n = search_string_in_dataframe( df, "Free Nig" )
image_path =n['url'].values[0]
display(HTML(url = image_path))

df_list = []
data = df.people.values

for json_string in data:
    if pd.isna(json_string):
        df_list.append(None)
    else:
        try:
            # First try direct JSON parsing
            parsed_data = json.loads(json_string)
        except json.JSONDecodeError:
            try:
                # Try fixing common issues
                json_string_fixed = json_string.replace("'", '"')
                parsed_data = json.loads(json_string_fixed)
            except json.JSONDecodeError:
                try:
                    # Last resort - use ast.literal_eval for Python-like syntax
                    parsed_data = ast.literal_eval(json_string)
                except (ValueError, SyntaxError) as e:
                    print(f"Failed to parse: {json_string}")
                    print(f"Error: {str(e)}")
                    parsed_data = None
        
        if parsed_data is not None:
            try:
                normalized_df = pd.json_normalize(parsed_data)
                df_list.append(normalized_df)
            except Exception as e:
                print(f"Normalization failed for: {parsed_data}")
                print(f"Error: {str(e)}")
                df_list.append(None)
        else:
            df_list.append(None)

# Store results
df['people_df'] = df_list

# Show first 2 successful parses
#print([x for x in df_list[:2] if x is not None])

# For multiple entries per row (if people_df contains multiple rows):
df['genders'] = df['people_df'].apply(lambda x: list(x['gender']) if isinstance(x, pd.DataFrame) and 'gender' in x.columns else [])

#print(df['genders'].value_counts())
# Create a mask for lists containing 'female' AND NOT containing 'male'
fem_mask = (
    df['genders'].apply(lambda x: 'female' in x if isinstance(x, list) else False) & 
    ~df['genders'].apply(lambda x: 'male' in x if isinstance(x, list) else False)
)

masc_mask = (
    ~df['genders'].apply(lambda x: 'female' in x if isinstance(x, list) else False) & 
    df['genders'].apply(lambda x: 'male' in x if isinstance(x, list) else False)
)

unknown_mask = (
    ~df['genders'].apply(lambda x: 'female' in x if isinstance(x, list) else False) & 
    ~df['genders'].apply(lambda x: 'male' in x if isinstance(x, list) else False) 
   
)
# Filter the DataFrame
fem_only_df = df[fem_mask]
male_only_df = df[masc_mask]
unkn_df = df[unknown_mask]

# Create the new column
df['male/female'] = [
    'female' if fem else 
    'male' if masc else 
    'unknown' 
    for fem, masc in zip(fem_mask, masc_mask)
]

print(f"There are {len(fem_only_df)} objects attributed to females and {len(male_only_df)} \
objects attributed to males. There are {len(unkn_df)} objects which are attributed to \
people of unknown gender. This give us {round(len(fem_only_df)/len(df)*100, 2)} percent female, and \
{round(len(male_only_df)/len(df)*100, 2)} male, {round(len(unkn_df)/len(df)*100, 2)} percent unknown or mixed gender attributed items.")

There are 5647 objects attributed to females and 151220 objects attributed to males. There are 88714 objects which are attributed to people of unknown gender. This give us 2.29 percent female, and 61.38 male, 36.01 percent unknown or mixed gender attributed items.

display(pd.DataFrame(df.genders.value_counts()))

plt.figure(figsize=(10, 6))

sns.histplot(data=male_only_df, x='dateend', kde=True, color = 'yellow', label = "male object created")
sns.histplot(data=unkn_df, x='dateend', kde=True, color="skyblue", label = "unknown object created")
sns.histplot(data=fem_only_df, x='dateend', kde=True, color = 'red', label = "fem object created")
plt.title('Gender Attributed to Objects by Year Created')
plt.xlabel('Year')
plt.ylabel('Frequency')
plt.legend(loc='upper left')
plt.grid(True)
plt.show()

plt.figure(figsize=(10, 6))

sns.histplot(data=male_only_df, x='dateend', kde=True, color = 'yellow', label = "male object created",bins=70)
sns.histplot(data=unkn_df, x='dateend', kde=True, color="skyblue", label = "unknown object created", bins = 270)
sns.histplot(data=fem_only_df, x='dateend', kde=True, color = 'red', label = "fem object created", bins= 60)
plt.title('Gender Attributed to Objects by Year Created')
plt.xlim(-1000, 1000)
plt.ylim(0, 4500)
plt.xlabel('Year')
plt.ylabel('Frequency')
plt.legend(loc='upper left')
plt.grid(True)
plt.show()


plt.figure(figsize=(10, 6))

sns.histplot(data=unkn_df, x='dateend', kde=True, color="skyblue", label = "unknown object created", bins = 500)
sns.histplot(data=male_only_df, x='dateend', kde=True, color = 'yellow', label = "male object created", bins = 120)
sns.histplot(data=fem_only_df, x='dateend', kde=True, color = 'red', label = "fem object created", bins = 100)
plt.title('Gender Attributed to Objects by Year Created')
plt.xlim(1000, 2050)
plt.ylim(0, 20000)
plt.xlabel('Year')
plt.ylabel('Frequency')
plt.legend(loc='upper left')
plt.grid(True)
plt.show()

recent_women = ((fem_only_df['dateend']<= 1975) & (fem_only_df['dateend']>= 1950)).sum()
recent_men = ((male_only_df['dateend']<= 1975) & (male_only_df['dateend']>= 1950)).sum()
recent_unknown = ((unkn_df['dateend']<= 1975) & (unkn_df['dateend']>= 1950)).sum()
total = recent_women + recent_men + recent_unknown
print(f"From objects made between 1950-1975 {round(recent_women/total *100,2)} percent were attributed to women while \
{round(recent_men/total *100,2)} percent were attributed to men and {round(recent_unknown/total *100,2)} percent were unknown.")

recent_women = ((fem_only_df['dateend']<= 2000) & (fem_only_df['dateend']>= 1975)).sum()
recent_men = ((male_only_df['dateend']<= 2000) & (male_only_df['dateend']>= 1975)).sum()
recent_unknown = ((unkn_df['dateend']<= 2000) & (unkn_df['dateend']>= 1975)).sum()
total = recent_women + recent_men + recent_unknown
print(f"From objects made between 1975-2000 {round(recent_women/total *100,2)} percent were attributed to women while \
{round(recent_men/total *100,2)} percent were attributed to men and {round(recent_unknown/total *100,2)} percent were unknown.")

recent_women = (fem_only_df['dateend']>= 2000).sum()
recent_men = (male_only_df['dateend']>= 2000).sum()
recent_unknown = (unkn_df['dateend']>= 2000).sum()
total = recent_women + recent_men + recent_unknown
print(f"From objects made between 2000-2025 {round(recent_women/total *100,2)} percent were attributed to women while \
{round(recent_men/total *100,2)} percent were attributed to men and {round(recent_unknown/total *100,2)} percent were unknown.")

From objects made between 1950-1975 2.34 percent were attributed to women while 77.0 percent were attributed to men and 20.66 percent were unknown.
From objects made between 1975-2000 11.99 percent were attributed to women while 76.29 percent were attributed to men and 11.72 percent were unknown.
From objects made between 2000-2025 18.79 percent were attributed to women while 58.65 percent were attributed to men and 22.56 percent were unknown.

gender_by_year = (
    df.groupby(['dateend', 'male/female'])
    .size()
    .unstack(fill_value=0)
)
gender_by_year = gender_by_year.div(gender_by_year.sum(axis=1), axis=0) * 100

# 3. Plot the results
plt.figure(figsize=(12, 5))

# Plot female percentage
plt.plot(gender_by_year.index, gender_by_year['female'], 
         label='Female', color='red', marker='o')

# Plot male percentage
plt.plot(gender_by_year.index, gender_by_year['male'], 
         label='Male', color='blue', marker='o')

# Formatting
plt.title('Gender Percentage of Objects in Collection by Year Created', fontsize=14)
plt.xlabel('Year', fontsize=12)
plt.ylabel('Percentage (%)', fontsize=12)
plt.legend(fontsize=12)
plt.grid(True, alpha=0.3)

# Rotate x-axis labels if many years
if len(gender_by_year) > 10:
    plt.xticks(rotation=45)
    plt.xlim(1900, 2026)

plt.tight_layout()
plt.show()

gender_by_year = (
    df.groupby(['accessionyear', 'male/female'])
    .size()
    .unstack(fill_value=0)
)
gender_by_year = gender_by_year.div(gender_by_year.sum(axis=1), axis=0) * 100

# 3. Plot the results
plt.figure(figsize=(12, 5))

# Plot female percentage
plt.plot(gender_by_year.index, gender_by_year['female'], 
         label='Female', color='red', marker='o')

# Plot male percentage
plt.plot(gender_by_year.index, gender_by_year['male'], 
         label='Male', color='blue', marker='o')

# Formatting
plt.title('Gender Percentage of Objects in Collection by Year Acquired', fontsize=14)
plt.xlabel('Year', fontsize=12)
plt.ylabel('Percentage (%)', fontsize=12)
plt.legend(fontsize=12)
plt.grid(True, alpha=0.3)

# Rotate x-axis labels if many years
if len(gender_by_year) > 10:
    plt.xticks(rotation=45)
    plt.xlim(1900, 2026)

plt.tight_layout()
plt.show()

df[df['dateend']>= 2020][['dateend', 'people_df', 'male/female']]

exhibited_df = df[df['exhibitioncount']>0]
gender_by_year = (
    exhibited_df.groupby(['dateend', 'male/female'])
    .size()
    .unstack(fill_value=0)
)
gender_by_year = gender_by_year.div(gender_by_year.sum(axis=1), axis=0) * 100

# 3. Plot the results
plt.figure(figsize=(12, 5))

# Plot female percentage
plt.plot(gender_by_year.index, gender_by_year['female'], 
         label='Female', color='red', marker='o')

# Plot male percentage
plt.plot(gender_by_year.index, gender_by_year['male'], 
         label='Male', color='blue', marker='o')

# Formatting
plt.title('Gender Percentage of Objects in Collection by Year Created (Eventually Exhibited)', fontsize=14)
plt.xlabel('Year', fontsize=12)
plt.ylabel('Percentage (%)', fontsize=12)
plt.legend(fontsize=12)
plt.grid(True, alpha=0.3)

# Rotate x-axis labels if many years
if len(gender_by_year) > 10:
    plt.xticks(rotation=45)
    plt.xlim(1900, 2026)

plt.tight_layout()
plt.show()

exhibited_df = df[df['exhibitioncount']>0]
gender_by_year = (
    exhibited_df.groupby(['accessionyear', 'male/female'])
    .size()
    .unstack(fill_value=0)
)
gender_by_year = gender_by_year.div(gender_by_year.sum(axis=1), axis=0) * 100

# 3. Plot the results
plt.figure(figsize=(12, 5))

# Plot female percentage
plt.plot(gender_by_year.index, gender_by_year['female'], 
         label='Female', color='red', marker='o')

# Plot male percentage
plt.plot(gender_by_year.index, gender_by_year['male'], 
         label='Male', color='blue', marker='o')

# Formatting
plt.title('Gender Percentage of Objects in Collection by Year Acquired (Eventually Exhibited)', fontsize=14)
plt.xlabel('Year', fontsize=12)
plt.ylabel('Percentage (%)', fontsize=12)
plt.legend(fontsize=12)
plt.grid(True, alpha=0.3)

# Rotate x-axis labels if many years
if len(gender_by_year) > 10:
    plt.xticks(rotation=45)
    plt.xlim(1900, 2026)

plt.tight_layout()
plt.show()

# For multiple entries per row (if people_df contains multiple rows):
df['artist_cultures'] = df['people_df'].apply(lambda x: list(x['culture']) if isinstance(x, pd.DataFrame) and 'culture' in x.columns else [])
df['artist_main_culture'] = df['artist_cultures'].apply(lambda x: x[0] if isinstance(x, list) and x != [] else [])

female_cultures = df[df['male/female'] == 'female']['artist_main_culture'].value_counts()
f_art_cultures = df[df['male/female'] == 'female']['culture'].value_counts()

plt.figure(figsize=(12,6))
plt.bar(height = female_cultures, x = female_cultures.index, label = 'Artist culture', alpha = .7)
plt.bar(height = f_art_cultures, x = f_art_cultures.index, label = 'Object culture', alpha = .7)
plt.xticks(rotation=90)
plt.title("Female Artist's Culture")
plt.legend()
plt.show()

exhibited_df.head(1)

df_list = []
data = exhibited_df.exhibitions.values

for json_string in data:
    if pd.isna(json_string):
        df_list.append(None)
    else:
        try:
            # First try direct JSON parsing
            parsed_data = json.loads(json_string)
        except json.JSONDecodeError:
            try:
                # Try fixing common issues
                json_string_fixed = json_string.replace("'", '"')
                parsed_data = json.loads(json_string_fixed)
            except json.JSONDecodeError:
                try:
                    # Last resort - use ast.literal_eval for Python-like syntax
                    parsed_data = ast.literal_eval(json_string)
                except (ValueError, SyntaxError) as e:
                    print(f"Failed to parse: {json_string}")
                    print(f"Error: {str(e)}")
                    parsed_data = None
        
        if parsed_data is not None:
            try:
                normalized_df = pd.json_normalize(parsed_data)
                df_list.append(normalized_df)
            except Exception as e:
                print(f"Normalization failed for: {parsed_data}")
                print(f"Error: {str(e)}")
                df_list.append(None)
        else:
            df_list.append(None)

# Store results
exhibited_df['exhibitions_df'] = df_list

exhibited_df['exhibitions_df'][:3]
exhibited_df.loc[:,'exhibitions_date'] = exhibited_df.exhibitions_df.apply(lambda x: list(x['begindate']) if isinstance(x, pd.DataFrame) and 'begindate' in x.columns else [])

m_exhibited = exhibited_df[exhibited_df['male/female'] == 'male']
f_exhibited = exhibited_df[exhibited_df['male/female'] == 'female']
u_exhibited = exhibited_df[exhibited_df['male/female'] == 'unknown']
m_list = []
f_list = []
u_list = []
from itertools import chain

m_list.append([i[:] for i in m_exhibited['exhibitions_date'].values])
m_list_flattened = list(chain.from_iterable(chain.from_iterable(m_list)))
m_list_flattened = pd.to_datetime(m_list_flattened, format='ISO8601').year.tolist()
f_list.append(i for i in f_exhibited['exhibitions_date'].values[:])
f_list_flattened = list(chain.from_iterable(chain.from_iterable(f_list)))
f_list_flattened = pd.to_datetime(f_list_flattened, format='ISO8601').year.tolist()
u_list.append(i for i in u_exhibited['exhibitions_date'].values[:])
u_list_flattened = list(chain.from_iterable(chain.from_iterable(u_list)))
u_list_flattened = pd.to_datetime(u_list_flattened, format='ISO8601').year.tolist()




plt.figure(figsize=(10, 6))


sns.histplot(m_list_flattened , kde=True, color = 'skyblue', label = "male object exhibited", bins = 130)
sns.histplot(f_list_flattened , kde=True, color="red", label = "female object exhibited")
#sns.histplot(u_list_flattened , kde=True, color="yellow", label = "unknown gender object exhibited")


plt.title('Objects Exhibited by Gender')
plt.xlim(1860, 2026)
#plt.ylim(0, 4500)
plt.xlabel('Year')
plt.ylabel('Frequency')
plt.legend(loc='upper left')
plt.grid(True)
plt.show()

e_df = pd.read_csv('exhibitions.csv')
e_df['begindate'] = pd.to_datetime(e_df['begindate'], errors = 'coerce')
e_df['year'] = e_df['begindate'].dt.year

plt.figure(figsize=(12, 6))
plt.title("Number of Exhibitions by Year")
sns.histplot(e_df['year'], bins = 300)
plt.show()

def extract_from_json(people_entry):
    try:
        if isinstance(people_entry, str):
            data = ast.literal_eval(people_entry)
            return next((item['name'] for item in data if 'name' in item), None)
        elif isinstance(people_entry, list):
            return next((item['name'] for item in people_entry if 'name' in item), None)
    except:
        return None

e_df['name'] = e_df['people'].apply(extract_from_json)

print(f"{round( e_df ['people'].isna().sum()/ e_df.shape[0], 2) *100} % of exhibitions do not have a curator listed.")

name_list = e_df['name'].value_counts()[(e_df['name'].value_counts()>=3)].index
list(name_list)
plt.figure(figsize=(12, 6))
for name in name_list:
    if name != None:
        j = e_df[e_df['name'] == name]['year']
        sns.histplot(j, label= name, bins = 30)
plt.xlim(1950, 2026)
plt.title("Exhibitions by Curators with Greater than 3 Exhibitions)")
plt.legend()
plt.show

Laura = e_df[e_df['name'] == 'Laura Muir']['year']
Stuart = e_df[e_df['name'] == 'Stuart Cary Welch']['year']
Robert = e_df[e_df['name'] == 'Robert D. Mowry']['year']
Mary = e_df[e_df['name'] == 'Mary McWilliams']['year']
Stephan = e_df[e_df['name'] == 'Stephan Wolohojian']['year']
Mary2 = e_df[e_df['name'] == 'Mary Schneider Enriquez']['year']
Henri = e_df[e_df['name'] == 'Henri Zerner']['year']
Susan = e_df[e_df['name'] == 'Susanne Ebbinghaus']['year']
Lyne = e_df[e_df['name'] == 'Lynette Roth']['year']
Amy = e_df[e_df['name'] == 'Amy Brauer  ']['year']
Mary2 = e_df[e_df['name'] == 'Mary Schneider Enriquez']['year']
Henri = e_df[e_df['name'] == 'Henri Zerner']['year']
Susan = e_df[e_df['name'] == 'Susanne Ebbinghaus']['year']
Lyne = e_df[e_df['name'] == 'Lynette Roth']['year']
Amy = e_df[e_df['name'] == 'Amy Brauer  ']['year']


plt.figure(figsize=(12, 6))
sns.histplot(Laura, label= "Laura Muir", bins = 30)
sns.histplot(Stuart, label= 'Stuart Cary Welch', bins = 30)
sns.histplot(Robert, label= 'Robert D. Mowry', bins = 30)
sns.histplot(Mary, label= 'Mary McWilliams', bins = 30)
sns.histplot(Stephan, label= 'Stephan Wolohojian', bins = 30)
sns.histplot(Mary2, label= 'Mary Schneider Enriquez', bins = 30)
sns.histplot(Henri, label= 'Henri Zerner', bins = 30)
sns.histplot(Susan, label= 'Susanne Ebbinghaus', bins = 30)
sns.histplot(Lyne, label= 'Lynette Roth', bins = 30)
sns.histplot(Amy, label= 'Amy Brauer  ', bins = 30)
#sns.histplot(Henri, label= 'Henri Zerner', bins = 30)
#sns.histplot(Susan, label= 'Susanne Ebbinghaus', bins = 30)
#sns.histplot(Lyne, label= 'Lynette Roth', bins = 30)
#sns.histplot(Amy, label= 'Amy Brauer  ', bins = 30)
plt.title("Exhibitions by Top 9 Curators by Frequency")
plt.legend()
plt.show()

93.0 % of exhibitions do not have a curator listed.

for person in name_list:
    shows= e_df[e_df['name'] == person][['title', 'year']]
    print(person)
    display(shows)

Stuart Cary Welch

Laura Muir

Robert D. Mowry

Mary McWilliams

Stephan Wolohojian

Mary Schneider Enriquez

Henri Zerner

Susanne Ebbinghaus

Lynette Roth

Amy Brauer

plt.figure(figsize=(12, 6))
sns.histplot(data=df[df['division']== 'Modern and Contemporary Art'], x='accessionyear', kde=True, label = 'Modern and Contemporary Art', bins = 70)
sns.histplot(data=df[df['division']== 'European and American Art'], x='accessionyear', kde=True, label = 'European and American Art', bins = 130)
sns.histplot(data=df[df['division']== 'Asian and Mediterranean Art'], x='accessionyear', kde=True, label = 'Asian and Mediterranean Art', bins = 100)
plt.title("Division of Object Acquired by Year")
plt.ylim(0,11000)
plt.xlim(1880,2026)
plt.legend()
plt.show()

top_20 = df['culture'].value_counts()[:20].index
plt.figure(figsize=(12, 6))
for cult in top_20: 
    sns.histplot(data=df[df['culture']== cult], x='accessionyear', label = cult, bins = 50)
plt.title("Culture of Object Acquired by Year")
plt.ylim(0,5000)
plt.xlim(1880,2026)
plt.legend()
plt.show()

top_8 = df['culture'].value_counts()[:8].index
plt.figure(figsize=(12, 6))
for cult in top_8: 
    sns.histplot(data=df[df['culture']== cult], x='accessionyear', label = cult, bins= 50)
plt.title("Culture of Object Acquired by Year")
plt.ylim(0,5000)
plt.xlim(1880,2026)
plt.legend()
plt.show()

des_df = e_df[~e_df['description'].isna()]

print(f"{round(des_df.shape[0]/e_df.shape[0], 2) *100}% of exhibitions have a description.")
des_df['begindate'] = pd.to_datetime(des_df['begindate'])
des_df['year'] = des_df['begindate'].dt.year
plt.figure(figsize=(12, 6))
plt.title("Exhibition With Descriptions by Year")
sns.histplot(des_df['year'], bins = 300)
plt.show()

5.0% of exhibitions have a description.

small_remove_words = set(['exhibition', 'harvard', 'art', 'work', 'university', 'museum',
'artists', 'museums', 'collection', 'works'])
remove_museum_words = set(['exhibition', 'harvard', 'art', 'work', 'university', 'museum',
'artists', 'museums', 'collection', 'works', 'organized', 'possible', 'made', 
                          'collections'])

colors = ['viridis', 'plasma', 'twilight_shifted', 'hsv','inferno', 'magma', 'cividis']
def print_wordclouds(df, remove):
    dft = {}
    for i in range(0, 30, 5):
       
        
        temp_df = df[(df['year'] >= (1995 + i)) & (df['year'] <= (2000 + i))]
        
        text = ' '.join(temp_df['description'].astype(str))
        text = text.lower()
        text = text.translate(str.maketrans('', '', string.punctuation))
        
        stop_words = set(stopwords.words('english'))
        stop_words = stop_words.union(remove)  
        word_tokens = word_tokenize(text)
        filtered_text = [word for word in word_tokens if word not in stop_words]
        
        clean_text = ' '.join(filtered_text)
        
        wordcloud = WordCloud(
            width=800, 
            height=400,
            background_color='white',
            max_words=100,
             colormap= colors[int(i/5)]
        ).generate(clean_text)
        
        plt.figure(figsize=(12, 8))
        plt.imshow(wordcloud, interpolation='bilinear')
        plt.title(f'Word Cloud for {1995+i}-{2000+i}')  # Added title for clarity
        plt.axis('off')
        plt.show()
        
        dft[f'des_{1995+i}'] = clean_text
    
    return dft
#dft = print_wordclouds(des_df, small_remove_words)
dft = print_wordclouds(des_df, remove_museum_words)

def print_wordclouds(df, remove):
    dft = {}
    for i in range(0,7, 1):
       
        
        temp_df = df[(df['year'] >= (2019 + i)) & (df['year'] <= (2020 + i))]
        
        text = ' '.join(temp_df['description'].astype(str))
        text = text.lower()
        text = text.translate(str.maketrans('', '', string.punctuation))
        
        stop_words = set(stopwords.words('english'))
        stop_words = stop_words.union(remove)  
        word_tokens = word_tokenize(text)
        filtered_text = [word for word in word_tokens if word not in stop_words]
        
        clean_text = ' '.join(filtered_text)
        
        wordcloud = WordCloud(
            width=800, 
            height=400,
            background_color='white',
            max_words=100,
            colormap= colors[i]
        ).generate(clean_text)
        
        plt.figure(figsize=(12, 8))
        plt.imshow(wordcloud, interpolation='bilinear')
        plt.title(f'Word Cloud for {2019+i}')  # Added title for clarity
        plt.axis('off')
        plt.show()
        
        dft[f'des_{2019+i}'] = clean_text
    
    return dft
#dft = print_wordclouds(des_df, small_remove_words)
dft = print_wordclouds(des_df, remove_museum_words)

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.cluster import KMeans
from sklearn.metrics.pairwise import cosine_similarity
import matplotlib.pyplot as plt
from sklearn.decomposition import PCA
from sklearn.metrics import silhouette_score

# 1. Preprocess text 
def preprocess(text, remove_words):
    text = text.lower()
    text = text.translate(str.maketrans('', '', string.punctuation))
    stop_words = set(stopwords.words('english')).union(remove_words)
    return ' '.join([word for word in word_tokenize(text) if word not in stop_words])


# 2. Vectorize with TF-IDF
vectorizer = TfidfVectorizer(max_features=1000, 
                            ngram_range=(1,2))  # Include bigrams
tfidf_matrix = vectorizer.fit_transform(des_df['description'].apply(preprocess, remove_words=remove_museum_words))

# 3. Cluster 
num_clusters = 36 #  adjust 
km = KMeans(n_clusters=num_clusters)
clusters = km.fit_predict(tfidf_matrix)
des_df['cluster'] = clusters 


print(silhouette_score(tfidf_matrix, clusters))

# 4. Visualize (using PCA to reduce dimensions)
pca = PCA(n_components=2)
reduced = pca.fit_transform(tfidf_matrix.toarray())
plt.scatter(reduced[:,0], reduced[:,1], c=clusters)
plt.legend()
plt.show()

# 5. Examine cluster keywords
print("Top terms per cluster:")
order_centroids = km.cluster_centers_.argsort()[:, ::-1]
terms = vectorizer.get_feature_names_out()
for i in range(num_clusters):
    print(f"Cluster {i}:")
    print(", ".join([terms[ind] for ind in order_centroids[i, :10]]))

# Store clusterd words list in data frame
cluster_keywords = {i: [terms[ind] for ind in order_centroids[i, :10]] 
                   for i in range(num_clusters)}

des_df['cluster_name'] = des_df['cluster'].map(cluster_keywords)

0.043813635974762045

Top terms per cluster:
Cluster 0:
2011, third, second, november, installation, october, feature selected, gallery installations, installation accompanies, courses feature
Cluster 1:
installation, objects coordinated, beever curator, coordinated amy, heath beever, heath, diane heath, brauer diane, course, amy brauer
Cluster 2:
calderwood, norma jean, norma, jean calderwood, islamic, jean, fabrics, mcwilliams, mary mcwilliams, mary
Cluster 3:
ancient, roman, vessels, portraits, greek, objects, munch, rothko, research, conservation
Cluster 4:
indian, islamic, india, islamic later, later indian, later, local, colonial, installation, period
Cluster 5:
design, school design, school, graduate school, modern, le, building, et, graduate, de
Cluster 6:
contemporary, modern contemporary, modern, body, wilmarth, pulitzer, fund modern, installation, installed, artist
Cluster 7:
students, artwork, course, humanities, analysis, faculty, architecture, world, visual, critical
Cluster 8:
lightbox, lightbox gallery, projects, digital, data, gallery, piene, sketchbooks, march, rivers
Cluster 9:
chinese, opium, scholar, 10, mowry, paintings, china, independent, robert, alan
Cluster 10:
buschreisinger, beuys, curator buschreisinger, walker, german, peter, artist, anniversary, presents, paintings
Cluster 11:
drawings, dutch, abrams, george abrams, george, catalogue, maida, william, maida george, van
Cluster 12:
european, european american, division european, american, prints, division, produced, called, paul, catalogue
Cluster 13:
african, institute, du, november, things, rotation, history, historical, course, college
Cluster 14:
african, installation, african american, american studies, gallery, fund, african african, arts, architecture, history
Cluster 15:
american, géricault, fogg, paintings, sculpture, field, includes, silver, curator american, william
Cluster 16:
buschreisinger, germany, german, object, vostell, modern, drawing, artist, medium, stage
Cluster 17:
photography, contemporary, photographs, curator photography, michelle lamunière, michelle, robinson family, barbara robinson, assistant, assistant curator
Cluster 18:
contemporary, schneider, fund, modern contemporary, modern, artist, pulitzer, fund modern, american, kent
Cluster 19:
sketchbooks, others, western, korea, east, every, japan, east asian, china, still
Cluster 20:
porcelain, european, natural, objects, 2006, buschreisinger, fellow, curatorial, chinese, became
Cluster 21:
social, images, photographs, prints, like, portraits, video, means, best, subjects
Cluster 22:
study, gallery, installation, persian, china, related, archaeology, boston, dedicated, center
Cluster 23:
islamic, roxburgh, calderwood curator, mcwilliams norma, mcwilliams, mary mcwilliams, norma, norma jean, calderwood, jean calderwood
Cluster 24:
chinese, arthur sackler, arthur, sackler, acquired, dworsky, curator chinese, alan dworsky, alan, tradition
Cluster 25:
rubens, greek, salt, might, great, stage, ancient, fragments, object, fogg
Cluster 26:
paintings, sculpture decorative, decorative, fogg, decorative arts, paintings sculpture, sculpture, arts, straus, degas
Cluster 27:
history, fund, installation, modern, pulitzer, history architecture, architecture, students, fund modern, department history
Cluster 28:
fogg, cooper, frances, curator modern, career, new, paintings, first, independent, modern
Cluster 29:
japanese, edo, prints, painting, japan, style, robert, fund, century, prince
Cluster 30:
bauhaus, 2019, school, february, fabrics, gropius, law, ii, buschreisinger, room
Cluster 31:
islamic, indian, india, south, later indian, islamic later, curator islamic, later, calderwood, norma jean
Cluster 32:
american, empire, colonial, century, john, spanish, thomas, catalogue, established, homer
Cluster 33:
color, fogg, paintings, artist, painting, bosch, louis, copley, curator contemporary, film
Cluster 34:
film, chamber, philosophy chamber, philosophy, science, scientific instruments, images, instruments, scientific, american
Cluster 35:
east, asian, east asian, chinese, months, symbols, korea, japan, dynasty, four

for cluster, group in des_df.groupby('cluster'):
    print(f"\n{group[['cluster_name']].values[0][0]}")
    display(group[['name', 'year', 'title']])  # Display as DataFrame

Cluster 0: 2011, third, second, november, installation, october, feature selected, gallery installations, installation accompanies, courses feature

Cluster 1: installation, objects coordinated, beever curator, coordinated amy, heath beever, heath, diane heath, brauer diane, course, amy brauer

Cluster 2: calderwood, norma jean, norma, jean calderwood, islamic, jean, fabrics, mcwilliams, mary mcwilliams, mary

Cluster 3: ancient, roman, vessels, portraits, greek, objects, munch, rothko, research, conservation

Cluster 4: indian, islamic, india, islamic later, later indian, later, local, colonial, installation, period

Cluster 5: design, school design, school, graduate school, modern, le, building, et, graduate, de

Cluster 6: contemporary, modern contemporary, modern, body, wilmarth, pulitzer, fund modern, installation, installed, artist

Cluster 7: students, artwork, course, humanities, analysis, faculty, architecture, world, visual, critical

Cluster 8: lightbox, lightbox gallery, projects, digital, data, gallery, piene, sketchbooks, march, rivers

Cluster 9: chinese, opium, scholar, 10, mowry, paintings, china, independent, robert, alan

# Get top terms per cluster
terms = vectorizer.get_feature_names_out()
order_centroids = km.cluster_centers_.argsort()[:, ::-1]


# Create a cluster label dictionary
cluster_labels = {}
for i in range(36):
    top_terms = [terms[ind] for ind in order_centroids[i, :10]]
    cluster_labels[i] = f"Cluster {i}: " + ", ".join(top_terms)
    print(cluster_labels[i])

# Add descriptive cluster names
des_df['cluster_name'] = des_df['cluster'].map(cluster_labels)

Cluster 0: 2011, third, second, november, installation, october, feature selected, gallery installations, installation accompanies, courses feature
Cluster 1: installation, objects coordinated, beever curator, coordinated amy, heath beever, heath, diane heath, brauer diane, course, amy brauer
Cluster 2: calderwood, norma jean, norma, jean calderwood, islamic, jean, fabrics, mcwilliams, mary mcwilliams, mary
Cluster 3: ancient, roman, vessels, portraits, greek, objects, munch, rothko, research, conservation
Cluster 4: indian, islamic, india, islamic later, later indian, later, local, colonial, installation, period
Cluster 5: design, school design, school, graduate school, modern, le, building, et, graduate, de
Cluster 6: contemporary, modern contemporary, modern, body, wilmarth, pulitzer, fund modern, installation, installed, artist
Cluster 7: students, artwork, course, humanities, analysis, faculty, architecture, world, visual, critical
Cluster 8: lightbox, lightbox gallery, projects, digital, data, gallery, piene, sketchbooks, march, rivers
Cluster 9: chinese, opium, scholar, 10, mowry, paintings, china, independent, robert, alan
Cluster 10: buschreisinger, beuys, curator buschreisinger, walker, german, peter, artist, anniversary, presents, paintings
Cluster 11: drawings, dutch, abrams, george abrams, george, catalogue, maida, william, maida george, van
Cluster 12: european, european american, division european, american, prints, division, produced, called, paul, catalogue
Cluster 13: african, institute, du, november, things, rotation, history, historical, course, college
Cluster 14: african, installation, african american, american studies, gallery, fund, african african, arts, architecture, history
Cluster 15: american, géricault, fogg, paintings, sculpture, field, includes, silver, curator american, william
Cluster 16: buschreisinger, germany, german, object, vostell, modern, drawing, artist, medium, stage
Cluster 17: photography, contemporary, photographs, curator photography, michelle lamunière, michelle, robinson family, barbara robinson, assistant, assistant curator
Cluster 18: contemporary, schneider, fund, modern contemporary, modern, artist, pulitzer, fund modern, american, kent
Cluster 19: sketchbooks, others, western, korea, east, every, japan, east asian, china, still
Cluster 20: porcelain, european, natural, objects, 2006, buschreisinger, fellow, curatorial, chinese, became
Cluster 21: social, images, photographs, prints, like, portraits, video, means, best, subjects
Cluster 22: study, gallery, installation, persian, china, related, archaeology, boston, dedicated, center
Cluster 23: islamic, roxburgh, calderwood curator, mcwilliams norma, mcwilliams, mary mcwilliams, norma, norma jean, calderwood, jean calderwood
Cluster 24: chinese, arthur sackler, arthur, sackler, acquired, dworsky, curator chinese, alan dworsky, alan, tradition
Cluster 25: rubens, greek, salt, might, great, stage, ancient, fragments, object, fogg
Cluster 26: paintings, sculpture decorative, decorative, fogg, decorative arts, paintings sculpture, sculpture, arts, straus, degas
Cluster 27: history, fund, installation, modern, pulitzer, history architecture, architecture, students, fund modern, department history
Cluster 28: fogg, cooper, frances, curator modern, career, new, paintings, first, independent, modern
Cluster 29: japanese, edo, prints, painting, japan, style, robert, fund, century, prince
Cluster 30: bauhaus, 2019, school, february, fabrics, gropius, law, ii, buschreisinger, room
Cluster 31: islamic, indian, india, south, later indian, islamic later, curator islamic, later, calderwood, norma jean
Cluster 32: american, empire, colonial, century, john, spanish, thomas, catalogue, established, homer
Cluster 33: color, fogg, paintings, artist, painting, bosch, louis, copley, curator contemporary, film
Cluster 34: film, chamber, philosophy chamber, philosophy, science, scientific instruments, images, instruments, scientific, american
Cluster 35: east, asian, east asian, chinese, months, symbols, korea, japan, dynasty, four

cluster_yearly = des_df.groupby(['year', 'cluster_name']).size().unstack(fill_value=0)

plt.figure(figsize=(12, 10))
cluster_yearly.plot.area(
    stacked=True,
    colormap='viridis',
    alpha=0.8
)
plt.title('Exhibition Clusters by Year')
plt.ylabel('Number of Exhibitions')
plt.xlabel('Year')
plt.legend(title='Cluster', bbox_to_anchor=(1.05, 1))
plt.tight_layout()
plt.show()

cluster_yearly.plot.bar(
    stacked=True,
    width=0.8,
    figsize=(12, 6),
    colormap='Paired'
)
plt.xticks(rotation=45)
plt.title('Yearly Cluster Distribution')
plt.legend(title='Cluster', bbox_to_anchor=(1.05, 1))
plt.show()

cluster_yearly.div(cluster_yearly.sum(axis=1), axis=0)\
   .plot.area(
       stacked=True,
       figsize=(12, 6),
       alpha=0.7
   )
plt.title('Cluster Proportions by Year')
plt.legend(title='Cluster', bbox_to_anchor=(1.05, 1))
plt.ylabel('Proportion')
plt.show()



plt.figure(figsize=(10, 8))
sns.heatmap(
    cluster_yearly.T,  # Transpose for clusters as rows
    annot=True,
    fmt='d',
    cmap='YlGnBu',
    linewidths=.5
)

plt.title('Exhibitions per Cluster per Year')
plt.ylabel('Cluster')
plt.xlabel('Year')
plt.show()

<Figure size 1200x1000 with 0 Axes>

plot_df = des_df[des_df['year'] != 1948].copy()
plot_df['name'] = plot_df['name'].fillna('')
plot_df['title'] = plot_df['title'].fillna('Untitled')  # Handle missing titles

# Create the visualization with additional hover data
fig = px.scatter(plot_df,
                 x='year',
                 y='cluster',
                 color='cluster',
                 hover_name='name',
                 hover_data={
                     'title': True,          # Add artwork titles
                     'cluster_name': True, 
                     'year': True, 
                     'cluster': False
                 },
                 title='Cluster Distribution Over Time with Keywords',
                 labels={'cluster': 'Cluster ID', 'year': 'Year'},
                 height=1200)


fig.update_traces(
    marker=dict(size=12, line=dict(width=1, color='DarkSlateGrey')),
    hovertemplate=(
        "Curator: <b>%{hovertext}</b><br>" +        
        "<i>%{customdata[0]}</i><br><br>" +  
        "Year: %{x}<br>" +
        "Cluster: %{y}<br>" +
        "Keywords: %{customdata[1]}<extra></extra>"
    )
)


fig.update_layout(
    xaxis=dict(tickmode='linear', dtick=5),  # Show ticks every 5 years
    yaxis=dict(tickvals=sorted(des_df['cluster'].unique())),
    hoverlabel=dict(
        bgcolor="white",
        font_size=12,
        font_family="Arial"
    )
)

fig.show()

e_df['clean_title'] = e_df['title'].apply(preprocess, remove_words=remove_museum_words)

# Load a lightweight model (good for titles)
#model = SentenceTransformer('all-MiniLM-L6-v2')  # 384-dim embeddings
#model = SentenceTransformer('paraphrase-multilingual-MiniLM-L12-v2') #multilingual
model = SentenceTransformer('all-mpnet-base-v2') # Higher quality (768-dim)

# Embed all titles
# Embed all titles
e_df['title_embedding'] = model.encode(e_df['clean_title'].tolist(), 
                                    convert_to_numpy=True).tolist()

embeddings = np.array(e_df['title_embedding'].tolist())                                
cos_dist = cosine_distances(embeddings)

clusterer = hdbscan.HDBSCAN(
    min_cluster_size=2,
    metric='precomputed', 
    cluster_selection_method='eom'
)

e_df['semantic_cluster_title'] = clusterer.fit_predict(cos_dist)


# 3. Visualize
reducer = umap.UMAP(metric='cosine')
projections = reducer.fit_transform(embeddings)
plt.scatter(projections[:,0], projections[:,1], c=e_df['semantic_cluster_title'], cmap='Spectral', s=10)
plt.legend()
plt.show()

from collections import defaultdict

# Get top titles per cluster
cluster_examples = defaultdict(list)
for clus in e_df['semantic_cluster_title'].unique():
    examples = e_df[e_df['semantic_cluster_title'] == clus]['clean_title'].sample(2, random_state=42)
    cluster_examples[clus] = examples.tolist()

for clus, examples in cluster_examples.items():
    print(f"\nCluster {clus}:")
    print("\n".join(f"- {ex}" for ex in examples))

Cluster 342:
- american photographs
- new american photographs

Cluster -1:
- private
- prints drawings bronzes degas

Cluster 173:
- 32q 1005 south arcade
- 32q 1003 west arcade

Cluster 170:
- 32q 3620 study gallery
- 32q 3610 teaching gallery

Cluster 1:
- harvardsmithsonian center astrophysics
- constellations light dark black white paper 19471954 sam francis

Cluster 134:
- haa 1 survey course s421 landmarks world architecture spring 2009
- landmarks world architecture

Cluster 246:
- haa 10 survey course s421 western tradition since renaissance fall 08 rotation 1
- haa 10 survey course s421 western tradition since renaissance fall 08 rotation 3

Cluster 36:
- review s425a small niche new acquisition lorna simpson
- review s426a large niche haa175w pop

Cluster 293:
- rocks mountains landscapes gardens essence east asian painting 1997
- chinese landscape painting

Cluster 83:
- recent acquisitions 2020
- recent acquisitions part ii building

Cluster 112:
- arts korea
- women arts asia

Cluster 352:
- chinese japanese painting early modern era
- later chinese japanese figure painting decorative arts

Cluster 171:
- 32q 3740 egyptian
- 32q 3440 middle east

Cluster 172:
- 32q 3700 roman
- 32q 1100

Cluster 148:
- 32q 2100 19th century
- 32q 2440 medieval

Cluster 166:
- 32q 1500 germany wars expressionisminterwar
- 32q 1440 secessionism munich vienna berlin expressionism

Cluster 143:
- 32q 2300 dutch flemish
- dutch 14501900

Cluster 353:
- modern german painting sculpture
- basic research selection postwar german painting sculpture

Cluster 104:
- george grosz
- robert kaufmann 19131959

Cluster 56:
- vija celmins
- prints vija celmins

Cluster 131:
- max beckmann america
- max beckmann

Cluster 27:
- selection important 19th century french masters aid central british fund jewish refugees rehabilitation
- exiles emigres 19331945

Cluster 7:
- discovery deceit archaeology forgers craft
- authentic fake

Cluster 269:
- jasper johns drawings
- drawings jasper johns

Cluster 62:
- early christian byzantine
- antiochontheorontes excavating early byzantine city

Cluster 291:
- paintings charles burchfield north midwest
- paintings charles burchfield charles sheeler

Cluster 15:
- notes harmonies nocturnes small james mcneill whistler
- notes harmonies nocturnes

Cluster 100:
- loan
- loan

Cluster 388:
- boston painters 18801930
- boston society watercolor painters

Cluster 122:
- celebration 75th anniversary year
- 20th anniversary

Cluster 119:
- 1776
- 1842

Cluster 120:
- 1993 biennial
- 1995 biennial

Cluster 392:
- french painting xixth xxth centuries
- 19th century french drawings

Cluster 390:
- american painting
- survey american painting

Cluster 125:
- 39th annual
- first annual

Cluster 37:
- memorabilia members
- collector

Cluster 349:
- loan paintings whittemore
- loan paintings claude monet

Cluster 78:
- agnes martin fifty small paintings pace
- agnes martin

Cluster 106:
- alexander james
- james brooks

Cluster 113:
- america west
- canada

Cluster 178:
- american women 18301930
- brilliant exiles american women paris 19001939

Cluster 391:
- american twentieth century painting sculpture
- 20th century abstract painting sculpture

Cluster 389:
- american watercolors 1880–1990 light
- water color paintings pastels drawings american european

Cluster 82:
- dürer ’ journeys travels renaissance artist
- american painter abroad frank duvenecks european years

Cluster 226:
- arts empire mughal india 15261705
- stories mughal india

Cluster 126:
- annual
- annual

Cluster 123:
- annual 1903
- annual 1862

Cluster 257:
- renaissance twelfth century
- antiquity renaissance

Cluster 45:
- coroplasts greek terracottas hellenistic world
- culture cycladic islands third millenium bc

Cluster 12:
- industrial
- international industry science

Cluster 23:
- goya time
- time

Cluster 303:
- 20th century ii cubism constructivism abstract
- cubism abstract

Cluster 261:
- new england silversmith
- arthur j stone 18471938 designer master silversmith

Cluster 90:
- loan ford madox brown preraphaelites
- collected ford madox brown

Cluster 50:
- winter
- winter

Cluster 165:
- ben shahn retrospective 18981969
- ben shahn retrospective

Cluster 287:
- unidentified allston club 1866
- unidentified algonquin club 1920

Cluster 334:
- expressionism
- birth impressionism

Cluster 20:
- blake room
- rooms

Cluster 252:
- boston society independent
- history boston library society

Cluster 151:
- canaletto
- venice canaletto rivals

Cluster 47:
- winter g f watts
- catalogue g f watts r

Cluster 212:
- catalogus der tentoonstelling van tekeningen uit de collectie van de heer van leeuwen te amsterdam
- collectie h van leeuwen te amsterdam tekeningen van 17e 19e eeuw

Cluster 304:
- joan waltemath recent paintings
- paintings miss cecilia beaux

Cluster 282:
- centenaire de cézanne
- centenaire du peintre independant paul cezanne

Cluster 375:
- paul cézanne
- paul cézanne

Cluster 289:
- charles sheeler retrospective
- charles sheeler

Cluster 192:
- chefs doeuvre de lécole française
- chefsdoeuvre de lart français

Cluster 311:
- christopher wilmarth drawing sculpture
- christopher wilmarth sculpture paper

Cluster 277:
- deuxieme exposition de peinture
- exposition de peinture française

Cluster 336:
- drawings claude lorrain
- drawings nicolas poussin claude lorrain magnasco society

Cluster 377:
- claude monet
- claude monet

Cluster 376:
- claude monet 18401926
- claude monet 18401926

Cluster 184:
- collages robert motherwell
- collages robert motherwell

Cluster 124:
- exposition doeuvres de géricault au profit de la societé la sauvegarde de lart français
- exposition doeuvres importantes de grands maîtres du dixneuvième siècle prêtées au profit de la cité universitaire de luniversité de paris

Cluster 10:
- business school student association kresge hall
- college association circulating

Cluster 142:
- david smith centennial
- david smith

Cluster 313:
- de delacroix à cézanne et van gogh
- exposition eugène delacroix peintures aquarelles pastels dessins gravures documents

Cluster 267:
- degas
- degas

Cluster 30:
- loan degas benefit new york infirmary
- degas loan benefit citizens committee children new york inc

Cluster 357:
- drawings degas
- drawings degas

Cluster 259:
- franz marc gedachtnisausstellung
- französische kunst des xix und xx jahrhunderts

Cluster 72:
- die sammlung oskar schmitz
- austellung oskar kokoschka

Cluster 310:
- unidentified kunsthaus zurich 1947
- werkbund bauhaus design germany 19001934

Cluster 203:
- paintings drawings preraphaelites circle
- drawings raphael circle british north american

Cluster 384:
- master drawings fogg
- special drawings fogg

Cluster 13:
- early science smith innovations instruments documents
- early science

Cluster 273:
- edvard munch
- edvard munch

Cluster 218:
- retrospective edward hopper
- edward hoppers new york

Cluster 317:
- eva hesse retrospective
- eva hesse

Cluster 140:
- c pissarro
- retrospective c pissarro

Cluster 297:
- william blake paintings watercolors drawings
- watercolors drawings william blake 17571827

Cluster 320:
- paintings watercolors john singer sargent ra
- drawings john singer sargent

Cluster 300:
- italian 12001900
- french 12001900

Cluster 236:
- ancient modern
- modern

Cluster 193:
- old masters
- five modern old masters

Cluster 319:
- pictures childe hassam
- pictures childe hassam

Cluster 231:
- pictures painted washington allston
- washington allston 17791843 loan paintings drawings memorabilia

Cluster 339:
- spanish painting
- spanish painting

Cluster 86:
- edward burnejones
- sir edward burnejones

Cluster 298:
- william blake
- william blake time

Cluster 157:
- william morris hunt
- mrs adele r levy memorial

Cluster 276:
- exposition decennale des beauxarts de 1889 1900
- exposition internationale

Cluster 275:
- exposition gericault peintre et dessinateur
- exposition géricault

Cluster 244:
- exposition ingres
- exposition ingres

Cluster 364:
- arab persian painting
- five masters persian indian painting

Cluster 258:
- frank stella paintings
- frank stella 1958

Cluster 301:
- french 19th century
- nineteenth century french

Cluster 346:
- gustave courbet
- ausstellung gustave courbet

Cluster 98:
- infancy old age gallery indians
- infancy old age gallery indians

Cluster 216:
- gaston lachaise
- gaston lachaise retrospective

Cluster 272:
- gauguin
- paul gauguin

Cluster 263:
- seurat paintings drawings
- pictures drawings georges seurat

Cluster 121:
- géricault 17911824
- mutiny géricault

Cluster 233:
- drawings principal paintings gorky
- drawings arshile gorky

Cluster 79:
- hans hartung
- hans hartung

Cluster 381:
- henri matisse retrospective paintings 18981939
- french drawings american clouet matisse

Cluster 69:
- collected w holman hunt om dcl
- collected w holman hunt om dcl

Cluster 16:
- national
- international

Cluster 238:
- sculpture international
- international fine arts british section

Cluster 204:
- jack levine
- david levine

Cluster 190:
- jacob lawrence struggle history american people
- jacob lawrence american struggle

Cluster 309:
- jacques lipchitz sculptor collector
- sculpture jacques lipchitz

Cluster 152:
- mr whistlers
- james mcneill whistler 1834 1903

Cluster 158:
- joan miró
- joan miró

Cluster 296:
- john la farge
- john la farge

Cluster 283:
- john singer sargent
- john singer sargent

Cluster 196:
- john singleton copley 17371815
- american john singleton copley

Cluster 278:
- la peinture française dans les américaines
- exposition de pastels français du xviie et du xviiie siècle

Cluster 80:
- latter days law images chinese buddhism 8501850
- saints shrines pilgrimages

Cluster 77:
- louise bourgeois
- suzanne valadon

Cluster 285:
- lyonel feininger
- lyonel feininger

Cluster 371:
- renoir
- renoir

Cluster 135:
- mark rothko paper
- mark rothko

Cluster 396:
- master drawings
- modern master drawings

Cluster 215:
- masterpieces
- masterpieces french

Cluster 130:
- max beckmann retrospective
- max beckmann retrospective

Cluster 210:
- niederländische meisterzeichnungen des 17 bis 19 jahrhunderts aus der sammlung hans van leeuwen
- niederländische zeichnungen des 1719 jahrhunderts aus der sammlung hans van leeuwen utrecht

Cluster 141:
- memorial paintings lilla cabot perry
- retrospective lilla cabot perry

Cluster 109:
- demuth memorial
- memorial raemisch

Cluster 359:
- paintings etchings lithographs james whistler
- etchings lithographs paintings drawings james mcneill whistler

Cluster 290:
- modern painting drawing sculpture collected louise joseph pulitzer jr
- modern painting drawing sculpture collected louise joseph pulitzer jr

Cluster 146:
- modernist idylls
- modernism designing new world

Cluster 374:
- monet
- monet

Cluster 107:
- moreau monticelli
- modigliani retrospective

Cluster 230:
- morris louis
- morris louis

Cluster 209:
- het nederlandsch stilleven
- hollandse tekeningen rond 1600

Cluster 84:
- new acquisitions 1957
- new acquisitions 1960

Cluster 93:
- new presences fogg
- new presences fogg

Cluster 85:
- gallery iv installation
- davis permanent gallery reinstallation

Cluster 294:
- renaissance florence 1470s
- 17th century europe

Cluster 247:
- éxposition des oeuvres de édouard manet
- oeuvres de manet

Cluster 400:
- old master drawings
- old master drawings

Cluster 132:
- old wedgwood bequest grenville lindall winthrop
- old wedgwood

Cluster 367:
- picasso
- picasso 1907

Cluster 350:
- paintings joseph h hazen
- paintings lent fine arts fogg

Cluster 373:
- paintings renoir
- renoir master impressionist

Cluster 331:
- paintings morris louis
- morris louis drawings 1948 1953

Cluster 347:
- private passion 19thcentury paintings drawings grenville l winthrop
- portrait prospect poetry british drawings grenville l winthrop bequest

Cluster 35:
- pavel tchelitchew
- pavel tchelitchew

Cluster 6:
- peter blume
- peter blume retrospect

Cluster 355:
- exposition de portraits du siècle
- exposition de portraits par ingres et ses élèves

Cluster 92:
- redon moreau bresdin
- prints rodolphe bresdin odilon redon

Cluster 66:
- raoul hausmann
- raoul dufy

Cluster 341:
- recent acquisitions photographs
- recent acquisitions part contemporary photographs

Cluster 386:
- renaissance baroque italian master drawings zuccari 15501600
- italian drawings michelangelo masters

Cluster 25:
- retrospective bazille
- retrospective

Cluster 81:
- revolution aesthetic practices 1917
- revolution tradition

Cluster 183:
- robert motherwell
- robert motherwell retrospective

Cluster 48:
- jubilee autumn
- royal jubilee

Cluster 239:
- salon dautomne
- salon dautomne

Cluster 99:
- sanford robinson gifford 18231880
- sanford robinson gifford 18231880

Cluster 127:
- sargents boston
- sargent sea

Cluster 155:
- sculpture kingdom benin
- bronzes ivories old kingdom benin

Cluster 385:
- selections dutch drawings maida george abrams
- seventeenthcentury dutch drawings selection maida george abrams

Cluster 60:
- sisley
- sisley

Cluster 95:
- sport american arts
- sport

Cluster 49:
- spring
- spring

Cluster 147:
- st louis
- st louis collects

Cluster 51:
- summer
- summer

Cluster 174:
- tamayo
- tamayo

Cluster 201:
- 8 americans 1964 venice bienniale
- unidentified venice biennale 1948

Cluster 281:
- american landscape
- landscapes

Cluster 299:
- van gogh
- van gogh america

Cluster 164:
- ben shahn
- ben shahn

Cluster 89:
- american 17501800 towards independence
- colonial america early republic

Cluster 270:
- complete graphic selection paintings paul gauguin
- gauguin paintings drawings prints sculpture

Cluster 144:
- hyman bloom retrospective
- hyman bloom matter life death

Cluster 38:
- friends collect
- members

Cluster 4:
- indian harvest part 1
- barbizon revisted

Cluster 76:
- maurice wertheim
- maurice wertheim

Cluster 362:
- maurice wertheim modern french artmonet picasso
- french paintings since 1870 maurice wertheim

Cluster 197:
- grand legacy arts ottoman empire
- ottomans contemporaries

Cluster 149:
- théodore gericault 17911824 bicentennial
- théodore géricault 17911824

Cluster 205:
- toulouselautrec
- toulouselautrec

Cluster 2:
- travelling
- travelling

Cluster 14:
- turner gainsborough
- gainsborough

Cluster 208:
- catalogus van de tentoonstelling van oude kunst uit het bezit van den internationalen handel
- tentoonstelling twee eeuwen engelsche kunst two centuries british

Cluster 21:
- fourth
- second

Cluster 332:
- unidentified mfa boston 1883
- unidentified mfa boston 1930

Cluster 255:
- unidentified new york historical society 1948
- unidentified brooklyn 1921

Cluster 321:
- unidentified manchester royal institution 1867
- unidentified royal academy arts

Cluster 337:
- unidentified boston athenaeum 1848
- unidentified boston athenaeum 1829

Cluster 182:
- unidentified lyon 1860
- unidentified marseilles 1860

Cluster 322:
- unidentified school fine arts 1941
- unidentified school fine arts boston 1934

Cluster 327:
- unidentified washington gallery 1961
- unidentified national gallery 1942

Cluster 312:
- salon 1799
- salon de 1853

Cluster 286:
- unidentified st botolph club 1894
- unidentified club new york 1956

Cluster 328:
- unidentified galerie bernheim jeune 1914
- unidentified galerie briant paris 1929

Cluster 254:
- unidentified institute chicago 1924
- unidentified institute modern 1947

Cluster 96:
- unidentified stedelijk 1905
- unidentified stedelijk 1945

Cluster 351:
- loan paintings winslow homer
- loan paintings winslow homer

Cluster 250:
- unidentified smith college 1944
- unidentified skidmore college 1942

Cluster 185:
- unidentified wadsworth atheneum 1948
- unidentified wadsworth atheneum 1935

Cluster 202:
- unidentified fogg 1958
- untitled fogg 1968

Cluster 253:
- unidentified san francisco 1937
- unidentified philadelphia 1936

Cluster 220:
- unidentified modern 1937
- unidentified wrul 1941

Cluster 103:
- unidentified faunce house brown 1933
- unidentified tuttle house 1938

Cluster 156:
- unidentified allen 1942
- unidentified allen memorial 1938

Cluster 110:
- unidentified brookline public library 1951
- unidentified lamont library 1959

Cluster 180:
- washington allston
- washington allston

Cluster 323:
- unidentified fine arts guild 1935
- unidentified fine arts associates 1952

Cluster 57:
- unidentified california palace legion honor 1954
- unidentified california palace legion honor 1931

Cluster 315:
- unidentified durlacher bros galleries 1958
- unidentified durlacher bros galleries 1951

Cluster 211:
- unidentified county st louis 1964
- unidentified los angeles county 1958

Cluster 222:
- unidentified lincoln laboratory 1986
- unidentified lincoln laboratory 1965

Cluster 302:
- rome eighteenth century
- venice eighteenth century

Cluster 114:
- benjamin franklin search better world
- benjamin franklin perspective

Cluster 344:
- winslow homer centenary
- memorial winslow homer

Cluster 345:
- winslow homer
- winslow homer

Cluster 264:
- meisterwerke griechischer kunst
- künstlerkopien

Cluster 70:
- unidentified davison center 1957
- unidentified davison center 1957

Cluster 232:
- gothic renaissance nuremberg 13001550
- triumph baroque architecture europe 16001750

Cluster 335:
- watercolors etchings edward hopper
- paintings watercolors etchings edward hopper

Cluster 368:
- water colors winslow homer
- water colors winslow homer

Cluster 369:
- watercolors winslow homer
- watercolors winslow homer 1836 1910

Cluster 340:
- winslow homer gloucester
- winslow homer new york state

Cluster 242:
- winslow homer critics forging national 1870s
- winslow homers subtropical america

Cluster 52:
- unidentified doll richards 1883
- unidentified doll richards 1888

Cluster 32:
- holidays
- christmas

Cluster 314:
- unidentified cambridge trust company 1951
- unidentified cambridge trust company 1956

Cluster 154:
- george bellows paintings drawings prints
- george bellows retrospective

Cluster 68:
- retrospective charles demuth
- charles demuth

Cluster 138:
- american
- american

Cluster 199:
- kunst und technik den 20er jahren neue sachlichkeit und gegenständlicher konstruktivismus
- ausstellung von kunstwerken des 15 bis 18 jahrhunderts aus basler privatbesitz

Cluster 305:
- rembrandt drawings american
- rembrandt america

Cluster 366:
- one hundred master drawings new england private
- master drawings drawings american private

Cluster 360:
- etchings lithographs lithotints drawings watercolors pastels childe hassam
- paintings childe hassam

Cluster 243:
- bill jensen drawings 1975 1981
- alfred jensen paintings diagrams years 19571977

Cluster 295:
- paintings studies sketches drawings mostly records travel 1886 1890 91 john la farge
- john la farge watercolors drawings

Cluster 370:
- drawing
- drawings

Cluster 241:
- french masterpieces 19th century henry p mcilhenny
- french masterpieces henry p mcilhenny

Cluster 74:
- lovis corinth retrospective
- lovis corinth

Cluster 292:
- nature charles burchfield memorial
- charles burchfield retrospective

Cluster 129:
- john marin
- john marin

Cluster 128:
- marin retrospective
- john marin retrospective

Cluster 234:
- drawings john ruskin 1819 1900
- drawings mr ruskin

Cluster 101:
- heinrich hoerle leben und werk 18951936
- kurt seligmann 19001962 leben und werk

Cluster 365:
- nineteenth century masterpieces
- masterpieces 17th 18th centuries

Cluster 266:
- jackson pollock drawings
- jackson pollock

Cluster 382:
- 11th international watercolors
- american watercolor society fifteenth annual

Cluster 393:
- modern drawings
- contemporary drawings

Cluster 111:
- fields david smith part ii
- fields david smith part iii

Cluster 372:
- delacroix cezanne french watercolors nineteenth century
- watercolors masters dürer cezanne

Cluster 288:
- david smith sculpture drawings paintings
- david smith formative years sculpture drawings 1930s 1940s

Cluster 397:
- 20th century master drawings
- nineteenth century master drawings

Cluster 73:
- line al hirschfeld retrospective
- hirschfeld century al hirschfeld retrospective

Cluster 195:
- islamic objects
- way favored objects world islam

Cluster 145:
- fragments antiquity drawing upon greek vases
- berlin painter world athenian vasepainting early fifth century

Cluster 223:
- john peter russell australian impressionist
- john leslie breck american impressionist

Cluster 177:
- scenes de la vie future images america european architecture 18931960
- sert half century architecture 19281979

Cluster 191:
- françois boucher north american 100 drawings
- drawings françois boucher

Cluster 189:
- degas dance
- degas ballet picturing movement

Cluster 167:
- portrait
- portrait

Cluster 159:
- titian tintoretto paolo veronese
- alberto giacometti

Cluster 262:
- sandro botticelli pittore della divina commedia
- piero di cosimo 14621522 painterpoet renaissance florence

Cluster 398:
- six centuries master drawings
- great master drawings seven centuries

Cluster 383:
- 20th international water colors
- eleventh annual new york water color club

Cluster 207:
- siqueiros pollock pollock siqueiros
- el greco pollock

Cluster 91:
- age elegance rococo effect
- age elegance rococo effect

Cluster 87:
- whistler
- whistler new england

Cluster 42:
- rico lebrun
- rico lebrun

Cluster 163:
- ben shahn graphics
- drawings ben shahn

Cluster 200:
- olieverfschetsen van rubens
- esquisses de rubens

Cluster 188:
- ed ruscha new paintings
- edward ruscha

Cluster 316:
- unidentified cambridge association 1948
- unidentified century association 1958

Cluster 206:
- portraits charles hopkinson
- portraits charles hopkinson

Cluster 387:
- american painting 17651963
- new england painting 17001775

Cluster 160:
- drawings beardsley period
- original drawings aubrey beardsley

Cluster 329:
- geometry spirit islamic illumination calligraphy
- traces calligrapher islamic calligraphy practice c 1600–1900

Cluster 271:
- salvador dalí rétrospective 19201980
- salvador dalí

Cluster 198:
- alexander calder retrospective
- alexander calder

Cluster 227:
- abstract surrealist american
- surrealism

Cluster 150:
- john white alexander memorial
- washington memorial

Cluster 251:
- gustave moreau 18261898
- gustave moreau

Cluster 139:
- jasper johns seasons
- jasper johns

Cluster 181:
- de verzameling van bernard houthakker
- de verzameling van schwarz

Cluster 306:
- unidentified boris mirski gallery 1950
- unidentified boris mirski gallery 1949

Cluster 58:
- three pleasant valley robert vonnoh oscar fehrer eugene higgins
- 4 amerikanare jasper johns alfred leslie robert rauschenberg richard stankiewicz

Cluster 162:
- mark rothko ’ murals
- mark rothko paintings paper

Cluster 326:
- master drawings woodner
- woodner selection drawings 1700

Cluster 380:
- paintings
- paintings

Cluster 228:
- margaret fisher drawings watercolors gouaches
- retrospective paintings stuart davis

Cluster 64:
- fragonard amoureux galant et libertin
- dominiquevivant denon loeil de napoléon

Cluster 399:
- old master drawings
- old master drawings

Cluster 379:
- drawn daily life dutch genre drawings maida george abrams
- things world selection dutch drawings maida george abrams

Cluster 333:
- master drawings exhibited bernard houthakker
- master drawings exhibited bernard houthakker 1972

Cluster 67:
- mrs herbert n straus
- mrs herbert n straus

Cluster 235:
- unto last two hundred years john ruskin
- ruskin

Cluster 168:
- maharaja splendour indias royal courts
- life court indias rulers

Cluster 44:
- paul klee
- paul klee loan american

Cluster 161:
- caspar david friedrich der graphiker handzeichnungen und radierungen
- paul klee lyonel feininger gemälde graphik xxix sonderausstellung

Cluster 0:
- balthus
- balthus

Cluster 394:
- master drawings 16001900
- master drawings 15001900

Cluster 279:
- old masters deceased masters british school including oil paintings miniatures drawings winter
- old masters deceased masters british school including water colour drawings c william blake frederick calvert samuel palmer louisa marchioness waterford

Cluster 308:
- baroque
- florentine baroque american

Cluster 17:
- paper sam reveles
- unique paper

Cluster 169:
- sébastien bourdon 16161671
- nicolas poussin

Cluster 354:
- loan drawings pastels edgar degas 18341917
- renoir degas loan drawings pastels sculptures

Cluster 358:
- edgar degas bronzes drawings pastels
- edgar degas paintings drawings pastels sculpture

Cluster 94:
- ballet
- dance

Cluster 219:
- visionary decade new voices 1940s boston
- 20th century new england

Cluster 361:
- prints drawings pieter bruegel elder
- pieter bruegel elder drawings prints

Cluster 153:
- sigmund abeles recent pastels
- sigmund abeles retrospective

Cluster 395:
- dutch flemish drawings 16th19th century
- 17th 18th century european drawings

Cluster 175:
- fiftieth anniversary europe xvithxviith centuries
- années 30 en europe 19291939

Cluster 117:
- rembrandt creates rembrandt ambition vision leiden 16291631
- rembrandt 16691969

Cluster 348:
- old master drawings chatsworth
- old master drawings chatsworth loan devonshire

Cluster 116:
- georg baselitz zeichnungen
- karl zerbe gouaches 1938

Cluster 325:
- unidentified ferargil galleries 1937
- unidentified vose galleries 1933

Cluster 118:
- gustav klimt landscapes
- gustav klimt 18621918 modernism making

Cluster 318:
- lyonel feininger watercolors drawings
- lyonel feininger oils watercolors 1940 1955

Cluster 88:
- blue four
- blue four

Cluster 225:
- american century culture 19002000 part 19001950
- american modernity 1865–1965

Cluster 26:
- anni albers
- anni albers

Cluster 108:
- henry g keller memorial
- memorial henry holt jr

Cluster 46:
- jan dibbets
- jan dibbets

Cluster 11:
- faculty
- teaching

Cluster 330:
- sixty paintings drawings peter paul rubens
- peter paul rubens drawings

Cluster 102:
- plum orchid chrysanthemum bamboo botanical motifs symbols east asian painting
- cultivating virtue botanical motifs symbols east asian

Cluster 65:
- male nude
- great american nude

Cluster 260:
- renaissance baroque bronzes alexis gregory
- renaissance modern bronzes

Cluster 63:
- william h lane foundation contemporary american
- twentieth century american william h lane foundation

Cluster 363:
- indian painting colonial period
- masters indian painting

Cluster 217:
- roy lichtenstein zeichnungen
- roy lichtenstein

Cluster 105:
- battistello caracciolo e il primo naturalismo napoli
- il camerino di alabastro antonio lombardo e la scultura allantica

Cluster 176:
- ernst ludwig kirchner mountain life
- ernst ludwig kirchner

Cluster 224:
- east meets west postwar prints drawings fromt two germanies
- two one printmaking germany 19451990

Cluster 55:
- new kind historical evidence photographs carpenter center
- carpenter center 40th anniversary

Cluster 59:
- poetry form richard tuttle drawings vogel
- richard tuttle retrospective

Cluster 307:
- eugène delacroix
- eugène delacroix 17981863

Cluster 187:
- eighteenth century european ceramics
- chinese ceramics one hundred selected masterpieces japan england france america

Cluster 284:
- presenting lyonel feininger retrospective minnesota
- lyonel feininger early friends

Cluster 22:
- sketches life west ireland
- sketches life west ireland

Cluster 256:
- walter richard sickert drawings
- retrospective drawings recent paintings walter richard sickert

Cluster 71:
- rome eternal city
- genius rome 15921623

Cluster 356:
- german realist drawings 1920s
- german realism twenties artist social critic

Cluster 28:
- cummings hilliard
- selections nathan cummings

Cluster 18:
- hokusai
- sakai hôitsu

Cluster 378:
- fine 19th 20th century british drawings watercolours
- romantic britain paintings drawings 17601860

Cluster 31:
- juan gris
- felix gonzaleztorres

Cluster 133:
- géricault cet inconnu 17911842 aquarelles gouaches dessins
- vijf eeuwen schilderijen aquarellen en tekeningen

Cluster 221:
- overlapping realms arts islamic world india 9001900
- review arts india islamic lands

Cluster 5:
- geometry motion leon polk smith paper
- moholynagy motion

Cluster 240:
- bauhaus modern
- bauhaus life

Cluster 61:
- review s228230 asian rotation 6
- review s424426 western 1560 1900

Cluster 34:
- british section international book industry graphic arts illustration books
- official catalogue

Cluster 29:
- review s426a large niche haa171x exoticism orientalism
- review s426a large niche 1 past present british 19th century

Cluster 248:
- haa 10 survey course s421 western tradition since renaissance fall 10 rotation 1
- haa 10 survey course s421 western tradition since renaissance fall 10 rotation 2

Cluster 97:
- zeichnungen alter meister aus deutschem privatbesitz
- handzeichnungen alter meister aus schweizer privatbesitz

Cluster 24:
- southern sung china
- korean united states

Cluster 249:
- haa 10 survey course s421 western tradition since renaissance fall 11 rotation 1
- haa 10 survey course s421 western tradition since renaissance fall 11 rotation 2

Cluster 338:
- annual boston athenaeum 1827
- boston athenaeum 1840

Cluster 237:
- fortythird annual pennsylvania academy fine arts
- eleventh annual pennsylvania academy fine arts

Cluster 229:
- landscape paintings edo period 250 years
- autumn leaves winter branches seasonal landscapes edo period

Cluster 3:
- japanese beauty rinpa
- treasures rinpa masters inheritance innovation

Cluster 19:
- jakuchu
- jakuchu another world

Cluster 115:
- national academy design annual
- national academy design annual 1857

Cluster 324:
- paintings late william morris hunt east gallery
- public sale celebrated paintings charcoal drawings late william morris hunt

Cluster 9:
- rachel whiteread shedding life
- rebecca horn “ progress ”

Cluster 53:
- poussin early years rome origins french classicism
- pomerance ancient

Cluster 194:
- fragonard
- fragonard

Cluster 213:
- nederlandse tekeningen der zestiende en zeventiende eeuw uit de collectie paul brandt
- nederlandse tekeningen der zestiende en zeventiende eeuw uit de collectie paul brandt

Cluster 268:
- lyonel feininger gemälde aquarelle zeichnungen 19071921
- lyonel feininger sonderausstellung seiner gemälde aquarelle zeichnungen und holzschnitte

Cluster 265:
- société de peintresgraveurs français quatrième exposition
- exposition de peintres et graveurs de paris

Cluster 136:
- jeanne mammen correct title unknown
- jeanne mammen retrospective

Cluster 40:
- willi baumeister zum gedächtnis
- willi baumeister malarstwo grafika

Cluster 41:
- willi baumeister
- willi baumeister

Cluster 280:
- old masters deceased masters british school winter
- old masters deceased masters british school winter

Cluster 214:
- werner heldt
- werner heldt retrospektive

Cluster 39:
- ruth asawa lifes
- ruth asawa line

Cluster 274:
- edvard munch 18631944
- edvard munch 18631944

Cluster 186:
- s424 indian southeast asian sculpture
- s425 east asian buddhist sculpture

Cluster 43:
- sarah reed blodgett portland
- paula modersohnbecker

Cluster 179:
- faith ringgold american people
- faith ringgold american people

Cluster 54:
- carl buchheister bilder und zeichnungen aus den 50er und 60er jahren
- die sammlung emil g bührle

Cluster 33:
- colloquium visual arts
- colloquium visual arts

Cluster 75:
- helvetica zeichnungen und graphik
- helvetica zeichnungen und graphik

Cluster 137:
- american british
- american british

Cluster 8:
- bill traylor
- derek hill

Cluster 343:
- contemporary photographs ii
- contemporary photographs vi

Cluster 245:
- dorothy herbert vogel
- selections dorothy herbert vogel

cluster_counts = e_df['semantic_cluster_title'].value_counts()
print(len(cluster_counts))
cluster_counts

402

semantic_cluster_title
-1      3234
 392      26
 123      24
 327      23
 173      16
 267      15
 390      15
 400      13
 250      13
 256      12
 389      12
 236      11
 334      11
 372      11
 288      11
 222      11
 295      11
 269      10
 259      10
 192      10
 367      10
 61       10
 76        9
 122       9
 148       9
 276       9
 283       9
 337       9
 131       9
 304       8
 100       8
 85        8
 375       8
 187       8
 125       8
 110       8
 83        8
 299       8
 310       8
 52        7
 378       7
 369       7
 335       7
 240       7
 330       7
 286       7
 157       7
 159       7
 80        7
 396       7
 374       7
 255       7
 107       7
 205       7
 201       7
 318       7
 360       7
 291       7
 349       7
 172       7
 352       7
 119       7
 272       6
 63        6
 273       6
 160       6
 307       6
 176       6
 386       6
 106       6
 399       6
 165       6
 278       6
 298       6
 72        6
 233       6
 169       6
 113       6
 11        6
 348       6
 7         6
 29        6
 243       6
 186       6
 104       6
 353       6
 182       6
 246       6
 212       6
 189       5
 345       5
 114       5
 37        5
 359       5
 344       5
 77        5
 135       5
 382       5
 370       5
 196       5
 296       5
 230       5
 185       5
 393       5
 202       5
 332       5
 394       5
 145       5
 350       5
 235       5
 328       5
 312       5
 322       5
 329       5
 190       5
 383       5
 144       5
 262       5
 152       5
 321       5
 346       5
 289       5
 357       5
 178       5
 142       5
 336       5
 277       5
 301       5
 284       5
 193       5
 338       5
 26        5
 115       5
 151       5
 121       5
 218       5
 381       5
 126       4
 297       4
 56        4
 317       4
 200       4
 149       4
 226       4
 4         4
 164       4
 384       4
 238       4
 62        4
 191       4
 204       4
 124       4
 177       4
 223       4
 68        4
 50        4
 314       4
 134       4
 368       4
 264       4
 128       4
 36        4
 33        4
 311       4
 73        4
 377       4
 254       4
 139       4
 229       4
 271       4
 87        4
 244       4
 373       4
 0         4
 388       4
 339       4
 279       4
 209       4
 395       4
 210       4
 216       4
 371       4
 285       4
 263       4
 79        4
 158       4
 16        4
 305       4
 15        4
 294       4
 161       4
 183       4
 347       4
 188       3
 111       3
 175       3
 42        3
 266       3
 234       3
 163       3
 162       3
 249       3
 133       3
 248       3
 74        3
 129       3
 217       3
 221       3
 40        3
 206       3
 241       3
 225       3
 137       3
 343       3
 58        3
 306       3
 75        3
 326       3
 280       3
 397       3
 167       3
 363       3
 44        3
 398       3
 333       3
 102       3
 9         3
 22        3
 19        3
 214       3
 324       3
 228       3
 39        3
 18        3
 198       3
 237       3
 41        3
 31        3
 195       3
 46        3
 354       3
 94        3
 153       3
 251       3
 342       3
 99        3
 239       3
 6         3
 355       3
 90        3
 341       3
 303       3
 12        3
 127       3
 35        3
 155       3
 385       3
 60        3
 95        3
 391       3
 49        3
 20        3
 331       3
 51        3
 215       3
 86        3
 275       3
 258       3
 69        3
 140       3
 309       3
 313       3
 252       3
 130       3
 141       3
 146       3
 84        3
 376       3
 282       3
 147       3
 231       3
 96        3
 281       3
 351       3
 293       3
 220       3
 143       3
 21        3
 180       3
 120       3
 1         3
 14        3
 170       3
 32        3
 57        3
 38        3
 154       3
 270       3
 138       3
 315       3
 171       3
 54        2
 356       2
 28        2
 71        2
 179       2
 53        2
 112       2
 30        2
 203       2
 184       2
 13        2
 8         2
 59        2
 55        2
 224       2
 320       2
 105       2
 10        2
 274       2
 43        2
 300       2
 194       2
 213       2
 268       2
 82        2
 265       2
 136       2
 78        2
 257       2
 45        2
 3         2
 47        2
 23        2
 261       2
 24        2
 97        2
 287       2
 27        2
 34        2
 166       2
 5         2
 199       2
 109       2
 319       2
 208       2
 181       2
 150       2
 227       2
 387       2
 316       2
 174       2
 91        2
 207       2
 89        2
 362       2
 197       2
 2         2
 253       2
 260       2
 103       2
 156       2
 323       2
 365       2
 101       2
 211       2
 302       2
 70        2
 292       2
 232       2
 340       2
 242       2
 48        2
 81        2
 25        2
 380       2
 65        2
 364       2
 108       2
 98        2
 88        2
 118       2
 325       2
 116       2
 117       2
 366       2
 361       2
 219       2
 290       2
 358       2
 17        2
 308       2
 93        2
 247       2
 132       2
 168       2
 67        2
 379       2
 92        2
 64        2
 66        2
 245       2
Name: count, dtype: int64

import plotly.express as px

top_clusters_df = e_df[e_df['semantic_cluster_title'].isin(cluster_counts.index[1:30])]


# 1. Prepare your data
plot_df = top_clusters_df.copy()
plot_df['name'] = plot_df['name'].fillna('')
plot_df['title'] = plot_df['title'].fillna('Untitled')  # Handle missing titles

# 2. Create the visualization with additional hover data
fig = px.scatter(plot_df,
                 x='year',
                 y='semantic_cluster_title',
                 color='semantic_cluster_title',
                 hover_name='name',
                 hover_data={
                     'title': True,          # Add artwork titles
                    
                     'year': True, 
                     'semantic_cluster_title': False
                 },
                 title='Cluster Distribution Over Time with Keywords (Clusters 1-30)',
                 labels={'cluster': 'Cluster ID', 'year': 'Year'},
                 height=900)

# 3. Customize hover template with title
fig.update_traces(
    marker=dict(size=12, line=dict(width=1, color='DarkSlateGrey')),
    hovertemplate=(
        "Curator: <b>%{hovertext}</b><br>" +        
        "Title: <i>%{customdata[0]}</i><br><br>" +  
        "Year: %{x}<br>" +
        "Cluster: %{y}<br>" 
       
    )
)

# 4. Improve layout
fig.update_layout(
    xaxis=dict(tickmode='linear', dtick=5),  # Show ticks every 5 years
    yaxis=dict(tickvals=sorted(des_df['cluster'].unique())),
    hoverlabel=dict(
        bgcolor="white",
        font_size=12,
        font_family="Arial"
    )
)

fig.show()



next_clusters_df = e_df[e_df['semantic_cluster_title'].isin(cluster_counts.index[30:60])]


# 1. Prepare your data
plot_df = next_clusters_df.copy()
plot_df['name'] = plot_df['name'].fillna('')
plot_df['title'] = plot_df['title'].fillna('Untitled')  # Handle missing titles

# 2. Create the visualization with additional hover data
fig = px.scatter(plot_df,
                 x='year',
                 y='semantic_cluster_title',
                 color='semantic_cluster_title',
                 hover_name='name',
                 hover_data={
                     'title': True,          # Add artwork titles
                    
                     'year': True, 
                     'semantic_cluster_title': False
                 },
                 title='Cluster Distribution Over Time with Keywords (Clusters 31-60)',
                 labels={'cluster': 'Cluster ID', 'year': 'Year'},
                 height=900)

# 3. Customize hover template with title
fig.update_traces(
    marker=dict(size=12, line=dict(width=1, color='DarkSlateGrey')),
    hovertemplate=(
        "Curator: <b>%{hovertext}</b><br>" +        
        "Title: <i>%{customdata[0]}</i><br><br>" +  
        "Year: %{x}<br>" +
        "Cluster: %{y}<br>" 
       
    )
)

# 4. Improve layout
fig.update_layout(
    xaxis=dict(tickmode='linear', dtick=5),  # Show ticks every 5 years
    yaxis=dict(tickvals=sorted(des_df['cluster'].unique())),
    hoverlabel=dict(
        bgcolor="white",
        font_size=12,
        font_family="Arial"
    )
)

fig.show()

import ast  # For safely evaluating strings
from itertools import chain

df_list = []
data = exhibited_df.people.values

for json_string in data:
    if pd.isna(json_string):
        df_list.append(None)
    else:
        try:
            # First try direct JSON parsing
            parsed_data = json.loads(json_string)
        except json.JSONDecodeError:
            try:
                # Try fixing common issues
                json_string_fixed = json_string.replace("'", '"')
                parsed_data = json.loads(json_string_fixed)
            except json.JSONDecodeError:
                try:
                    # Last resort - use ast.literal_eval for Python-like syntax
                    parsed_data = ast.literal_eval(json_string)
                except (ValueError, SyntaxError) as e:
                    print(f"Failed to parse: {json_string}")
                    print(f"Error: {str(e)}")
                    parsed_data = None
        
        if parsed_data is not None:
            try:
                normalized_df = pd.json_normalize(parsed_data)
                df_list.append(normalized_df)
            except Exception as e:
                print(f"Normalization failed for: {parsed_data}")
                print(f"Error: {str(e)}")
                df_list.append(None)
        else:
            df_list.append(None)

# Store results
exhibited_df['people_df'] = df_list
exhibited_df.loc[:,'artist'] = exhibited_df.people_df.apply(lambda x: list(x['name']) if isinstance(x, pd.DataFrame) and 'name' in x.columns else [])
exhibited_df.loc[:,'exhibition_dates'] = exhibited_df.exhibitions_df.apply(lambda x: list(x['begindate']) if isinstance(x, pd.DataFrame) and 'begindate' in x.columns else [])
from ast import literal_eval

def get_years(row):
    """Process dates without accidental character-level parsing"""
    dates = []
    for item in row['exhibition_dates']:
        if isinstance(item, list):
            dates.extend(item)  # Flatten nested lists
        else:
            dates.append(item)  # Add single date
    
    years = []
    for date_str in dates:
        try:
            # Handle various date formats
            if pd.notna(date_str) and str(date_str).strip():
                year = pd.to_datetime(date_str, errors='coerce').year
                if not pd.isna(year):
                    years.append(year)
        except Exception as e:
            print(f"Couldn't parse {date_str}: {e}")
    return years

   
exhibited_df.loc[:,'exhibition_years'] = exhibited_df.apply(get_years, axis = 1)


exhibited_df.loc[:,'artist_primary']= exhibited_df['artist'].apply(lambda x: x[0] if x != [] else []) 

artists_exhibited_counts = exhibited_df.artist_primary.value_counts()



artists_exhibited_counts[2:30].plot.bar(
  

    figsize=(14, 6),
   
)
plt.xticks(rotation=85)
plt.title('Number of Times an Artists Unique Work was Exhibited')

plt.show()

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[83], line 5
      2 from itertools import chain
      4 df_list = []
----> 5 data = exhibited_df.people.values
      7 for json_string in data:
      8     if pd.isna(json_string):

NameError: name 'exhibited_df' is not defined

# looking at the AWS face recognition enteries
aws_df = pd.read_csv('aws_face_annotations.csv')

# Extract Age
aws_df['age'] = aws_df['body'].str.extract(r'age: (\d+-\d+)')

# Extract Gender
aws_df['gender'] = aws_df['body'].str.extract(r'gender: (\w+)\(?')
aws_df['gender_confidence'] = aws_df['body'].str.extract(r'gender: \w+\((\d+\.\d+)%\)').astype(float)

# Extract emotions 
all_emotions = []
for idx, row in aws_df.iterrows():
    matches = re.findall(r'([A-Z]+): (\d+\.\d+)%', row['body'])
    for emotion, percent in matches:
        all_emotions.append({
            'original_index': idx,
            'emotion': emotion,
            'percentage': float(percent)
        })

# Create emotion DataFrame
emotion_df = pd.DataFrame(all_emotions)

# Pivot using groupby to avoid duplicates 
emotion_wide = (emotion_df.groupby(['original_index', 'emotion'])
                ['percentage']
                .first()  # Takes first if duplicates exist
                .unstack()
                .astype(float))

# Join with original data
aws_df = aws_df.join(emotion_wide)



aws_df['first_emotion'] = aws_df['body'].str.extract(r'<br/>([A-Z]+): \d+\.\d+%')


aws_df['raw_dict'] = aws_df['raw'].apply(ast.literal_eval)

def classify_gaze(raw, confidence_threshold=40):
    """Robust gaze classification with comprehensive error handling"""
    # Handle missing/invalid data cases
    if pd.isna(raw):
        return 'bad_data'
        
    try:
        # Handle both string and dict input like safe_extract
        if isinstance(raw, str):
            try:
                data = json.loads(raw)
            except json.JSONDecodeError:
                return 'bad_data'
        else:
            data = raw
            
        # Safely access EyeDirection with nested get()
        eye_data = data.get('EyeDirection', {})
        if not eye_data:  # Empty dict means no eye data
            return 'bad_data'
            
        # Extract values with defaults
        conf = eye_data.get('Confidence', 0)
        yaw = abs(eye_data.get('Yaw', 0))
        
        # Classification logic
        if conf < confidence_threshold:
            return 'low_conf'
        elif yaw < 10:
            return 'straight_gaze'
        elif 10 <= yaw <= 20:
            return 'side_gaze'
        else:
            return 'no_gaze'
            
    except Exception as e:  # Catch any unexpected errors
        print(f"Error processing gaze data: {e}")
        return 'bad_data'

# Apply to DataFrame
aws_df['gaze_classification'] = aws_df['raw_dict'].apply(classify_gaze)


#aws_df['smile'] = aws_df['raw'].apply(lambda x: x['Smile']['Value'])
#aws_df['MouthOpen'] = aws_df['raw'].apply(lambda x: x['MouthOpen']['Value'])
#aws_df['EyesOpen'] = aws_df['raw'].apply(lambda x: x['EyesOpen']['Value'])

def safe_extract(row, primary_key, secondary_key, default=None):
    """Safely extract nested dictionary values"""
    if pd.isna(row):
        return default
        
    try:
        # Handle both dict and JSON string cases
        if isinstance(row, str):
            data = json.loads(row)
        else:
            data = row
            
        return data.get(primary_key, {}).get(secondary_key, default)
    except (AttributeError, KeyError, json.JSONDecodeError):
        return default

# Apply to all features
aws_df['smile'] = aws_df['raw_dict'].apply(lambda x: safe_extract(x, 'Smile', 'Value', False))
aws_df['MouthOpen'] = aws_df['raw_dict'].apply(lambda x: safe_extract(x, 'MouthOpen', 'Value', False))
aws_df['EyesOpen'] = aws_df['raw_dict'].apply(lambda x: safe_extract(x, 'EyesOpen', 'Value', False))


male_df = aws_df[aws_df.gender=='Male']
female_df = aws_df[aws_df.gender=='Female']


from itertools import chain

# Sample Data

# Step 1: Convert ranges to lists of ages
aws_df['ages'] =aws_df['age'].apply(
    lambda x: list(range(int(x.split('-')[0]), int(x.split('-')[1]) + 1))
)


male_df = aws_df[aws_df.gender=='Male']
female_df = aws_df[aws_df.gender=='Female']
num_m = round(len(male_df)/len(aws_df)*100,2)
num_f = round(len(female_df)/len(aws_df)*100,2)


# Step 2: Flatten all ages
male_ages = list(chain.from_iterable(male_df['ages']))
female_ages =  list(chain.from_iterable(female_df['ages']))

plt.pie([num_m, num_f], labels=['male','female'], autopct='%1.1f%%')
plt.title("Face Gender Distribution in Art")
plt.show()

# Step 3: Plot
bins = np.arange(min(male_ages + female_ages), max(male_ages + female_ages) + 2)
plt.hist(male_ages, bins=bins, edgecolor='black', label = 'male', alpha=0.9)
plt.xlabel('Age')
plt.ylabel('Count')
plt.title('Artwork Age Distribution by Gender')

plt.hist(female_ages, bins=bins, edgecolor='black', label = 'female', alpha=0.7)
plt.legend()
plt.show()

print(f"Average female age is: {round(np.mean(female_ages), 2)}. \n Average male age is: {round(np.mean(male_ages),2)}")
fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(10, 5))
axes[0].set_title(f"Male Ages")
sns.violinplot(male_ages, label = 'male', ax=axes[0])
axes[1].set_title(f"Female Ages")
sns.violinplot(female_ages, label = 'female', ax=axes[1])
plt.show()
fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(10, 5))
sns.boxplot(male_ages, label = 'male', ax=axes[0])
sns.boxplot(female_ages, label = 'female', ax=axes[1])
plt.show()

Average female age is: 31.01. 
 Average male age is: 36.69

sns.histplot(data=male_df.reset_index(), x='first_emotion', stat='percent', discrete=True, label = 'male')
sns.histplot(data=female_df.reset_index(), x='first_emotion', stat='percent', discrete=True, label ='female')
plt.title('Primary Emotion Distribution (%)')
plt.xlabel('Emotion')
plt.ylabel('Percentage')
plt.legend()
plt.xticks(rotation=45)
plt.show()

emotion_cols = ['ANGRY', 'CALM', 'CONFUSED', 'DISGUSTED', 'FEAR', 'HAPPY',	'SAD', 'SURPRISED']
emotion_dict = {}
med_emotion_dict={}

for col in emotion_cols:
    print(col, male_df[col].median(), male_df[col].mean(),  female_df[col].median(), male_df[col].mean())
    
    emotion_dict.update({col:[ male_df[col].mean()/100,female_df[col].mean()/100] })
    med_emotion_dict.update({col:[ male_df[col].median()/100,female_df[col].median()/100] })

emotion_df = pd.DataFrame(emotion_dict, index = ['Male Mean', 'Female Mean'])
median_emotion_df = pd.DataFrame(med_emotion_dict, index = ['Male Median', 'Female Median'])
display(emotion_df)
display(median_emotion_df)
sns.barplot(x = emotion_df.columns, y = emotion_df.iloc[0], label = 'Male', alpha = .7)
sns.barplot(x = emotion_df.columns, y = emotion_df.iloc[1], label = 'Female', alpha = .7)
plt.xticks(rotation=45)
plt.ylabel('Percent Emotion')
plt.show()

ANGRY 1.928045 14.126399745477725 2.64739 14.126399745477725
CALM 52.332598 56.591701713678844 49.550758 56.591701713678844
CONFUSED 1.889152 13.699250638435256 2.216311 13.699250638435256
DISGUSTED 0.9813215 12.264021829467115 1.5310025 12.264021829467115
FEAR 2.213297 10.360356056744786 2.1639345 10.360356056744786
HAPPY 1.798896 18.138128892144078 9.815443 18.138128892144078
SAD 4.747918 20.740359067283464 11.6978625 20.740359067283464
SURPRISED 4.9772525000000005 13.942812918596145 6.324136 13.942812918596145

fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(12, 5))
axes[0].bar(height = [male_df['smile'].value_counts()[1]/len(male_df)*100,female_df['smile'].value_counts()[1]/len(female_df)*100] ,
        x = ['male', 'female'])
axes[0].set_title("Percent Smiling")
axes[0].set_ylim(0, 30)
axes[1].bar(height = [male_df['MouthOpen'].value_counts()[1]/len(male_df)*100,female_df['MouthOpen'].value_counts()[1]/len(female_df)*100] ,
        x = ['male', 'female'])
axes[1].set_title("Percent Mouth Open")
axes[1].set_ylim(0, 30)
plt.show()
plt.bar(height = [male_df['EyesOpen'].value_counts()[0]/len(male_df)*100,female_df['EyesOpen'].value_counts()[0]/len(female_df)*100] ,
        x = ['male', 'female'])
plt.title("Percent Eyes Open")
plt.ylim(0, 100)
plt.show()
gaze =['no_gaze', 'straight_gaze', 'side_gaze'] 
print(male_df['gaze_classification'].value_counts())
print(female_df['gaze_classification'].value_counts())
m_gaze = male_df['gaze_classification'][male_df['gaze_classification'].isin(gaze)].value_counts()
f_gaze = female_df['gaze_classification'][female_df['gaze_classification'].isin(gaze)].value_counts()
plt.bar(height = 100* m_gaze/len(male_df['gaze_classification'][male_df['gaze_classification'].isin(gaze)]), 
        x = m_gaze.index, alpha = .7, label = 'male')
plt.bar(height = 100*f_gaze/len(female_df['gaze_classification'][female_df['gaze_classification'].isin(gaze)]), 
        alpha =.7, x = f_gaze.index, label = 'female')
plt.title("Percent Gazing by Gender")
plt.legend()
plt.show()

gaze_classification
bad_data         251413
no_gaze             241
low_conf            160
straight_gaze       153
side_gaze           124
Name: count, dtype: int64
gaze_classification
bad_data         187707
low_conf            139
no_gaze             136
straight_gaze       125
side_gaze           102
Name: count, dtype: int64

	count
genders
[male]	126139
[unknown]	52186
[]	31393
[male, male]	11114
[male, unknown]	7715
[female]	5162
[unknown, unknown]	3127
[unknown, male]	2954
[unidentified]	1226
[male, male, male]	548
[male, unknown, unknown]	479
[female, male]	421
[male, unknown, male]	401
[male, male, unknown]	388
[unknown, unknown, unknown]	335
[female, unknown]	329
[unidentified, male]	322
[unknown, male, male]	314
[unidentified, unknown]	278
[male, female]	160
[unknown, unknown, male]	125
[unknown, male, unknown]	121
[unknown, unknown, unknown, unknown]	120
[unknown, male, unknown, unknown]	106
[unidentified, male, female]	83
[male, unidentified]	65
[female, female]	59
[male, male, unknown, unknown]	59
[male, male, unidentified]	57
[male, unknown, unknown, unknown]	41
[unknown, female]	36
[male, male, male, male]	35
[female, unknown, male]	30
[female, unknown, unknown]	28
[unknown, unidentified]	26
[male, male, unknown, male]	24
[unknown, male, unknown, male]	18
[female, male, unknown]	18
[male, unknown, male, male]	17
[male, female, unknown]	17
[male, unidentified, unknown]	14
[unknown, unknown, unknown, male]	13
[unknown, unknown, male, unknown]	12
[female, male, male]	12
[female, unknown, unknown, unknown]	12
[male, male, unknown, female]	12
[female, male, unidentified]	10
[unidentified, male, unknown]	10
[unknown, male, male, male]	9
[unidentified, unknown, unknown]	9
[male, unknown, unknown, male]	9
[male, unknown, male, unknown]	9
[unknown, male, unidentified]	9
[male, male, male, unknown]	8
[unidentified, male, male]	8
[unknown, male, male, unknown]	8
[unknown, unknown, male, male]	7
[female, unidentified]	7
[unidentified, unidentified]	7
[male, female, male]	6
[female, female, unknown]	6
[unidentified, unknown, male]	5
[unknown, male, male, unidentified]	5
[unknown, male, female]	5
[male, male, male, male, male]	4
[male, male, unidentified, male]	4
[male, unidentified, male]	4
[unknown, female, male]	3
[unidentified, unidentified, male, male]	3
[male, male, unknown, unidentified, female, male, unidentified, male, male]	2
[male, male, unidentified, male, male]	2
[unknown, male, male, unknown, male]	2
[female, female, male]	2
[male, unknown, unidentified]	2
[female, unknown, male, unknown]	2
[unknown, unidentified, unknown]	2
[male, male, male, male, male, male]	2
[unknown, unknown, unknown, unknown, unknown, unknown, unknown]	2
[unknown, unknown, unknown, unknown, unknown]	2
[unknown, female, unknown]	2
[male, male, unknown, unknown, unknown]	2
[unidentified, female]	2
[female, male, male, unidentified]	2
[unidentified, male, unknown, unidentified]	1
[unknown, unknown, male, male, male, male, male, male, male]	1
[unknown, unidentified, male]	1
[female, unidentified, unknown]	1
[unknown, male, unknown, male, male, male, male, male]	1
[male, unidentified, male, unknown]	1
[female, male, male, male]	1
[female, male, male, female]	1
[female, unknown, unknown, female]	1
[unidentified, female, male]	1
[unknown, unknown, unknown, unknown, male]	1
[unknown, unknown, unknown, male, unknown]	1
[unknown, unidentified, male, unidentified, male, male, unknown, male, unknown, male, unknown, unknown, unidentified, male, male, male, male]	1
[male, unknown, unidentified, unknown]	1
[male, male, male, male, unknown]	1
[female, unidentified, male]	1
[unknown, unidentified, male, male, male, unidentified, unidentified, unknown, male, male, male, male, male, male, unknown, male]	1
[unidentified, male, unknown, unknown]	1
[male, male, male, male, male, male, male, male, male, unknown]	1
[unknown, male, unknown, male, male, male]	1
[male, male, male, male, male, male, male, male]	1
[unknown, unknown, female]	1
[male, male, male, male, male, male, male, male, male]	1
[unidentified, unknown, unknown, unknown]	1
[male, male, male, unknown, unknown]	1
[male, male, unknown, unknown, male]	1
[male, male, unknown, male, unknown]	1
[male, male, male, female]	1
[male, male, female, unknown]	1
[unknown, male, unknown, female, male]	1
[unknown, unknown, male, unknown, male, unknown, unknown, unknown]	1
[unidentified, male, unidentified, unknown]	1
[female, male, male, male, unknown]	1
[male, male, unknown, unidentified]	1
[unknown, male, male, female, male, male, male, male, male, male, male, male, male, male, male]	1
[male, female, unknown, unknown]	1
[female, female, male, unidentified]	1
[female, unknown, unidentified]	1
[male, unidentified, unidentified]	1
[male, unknown, male, unidentified]	1
[male, male, male, male, male, unidentified]	1
[male, male, unidentified, unknown]	1
[unknown, unknown, male, unknown, unknown]	1

	dateend	people_df	male/female
82367	2020	role birthplace gender displaydate prefix culture \ 0 Artist Ilé-If?`, Nigeria female born 1985 None American displayname alphasort name personid \ 0 Toyin Ojih Odutola Ojih Odutola, Toyin Toyin Ojih Odutola 68282 deathplace displayorder 0 None 1	female
82368	2020	role birthplace gender displaydate prefix culture \ 0 Artist Ilé-If?`, Nigeria female born 1985 None American displayname alphasort name personid \ 0 Toyin Ojih Odutola Ojih Odutola, Toyin Toyin Ojih Odutola 68282 deathplace displayorder 0 None 1	female
82369	2020	role birthplace gender displaydate prefix culture \ 0 Artist Ilé-If?`, Nigeria female born 1985 None American displayname alphasort name personid \ 0 Toyin Ojih Odutola Ojih Odutola, Toyin Toyin Ojih Odutola 68282 deathplace displayorder 0 None 1	female
82521	2020	role birthplace gender displaydate prefix culture displayname \ 0 Artist San Francisco, CA male born 1969 None American Will Wilson alphasort name personid deathplace displayorder 0 Wilson, Will Will Wilson 68321 None 1	male
83270	2020	role birthplace gender displaydate prefix culture displayname \ 0 Artist None unknown born 1971 None None Hong Chun Zhang 張春紅 alphasort name personid deathplace displayorder 0 Zhang, Hong Chun Hong Chun Zhang 張春紅 68583 None 1	unknown
83271	2020	role birthplace gender displaydate prefix culture displayname \ 0 Artist None unknown born 1971 None None Hong Chun Zhang 張春紅 alphasort name personid deathplace displayorder 0 Zhang, Hong Chun Hong Chun Zhang 張春紅 68583 None 1	unknown
83272	2020	role birthplace gender displaydate prefix culture displayname \ 0 Artist None unknown born 1971 None None Hong Chun Zhang 張春紅 alphasort name personid deathplace displayorder 0 Zhang, Hong Chun Hong Chun Zhang 張春紅 68583 None 1	unknown
83273	2020	role birthplace gender displaydate prefix culture displayname \ 0 Artist None unknown born 1971 None None Hong Chun Zhang 張春紅 alphasort name personid deathplace displayorder 0 Zhang, Hong Chun Hong Chun Zhang 張春紅 68583 None 1	unknown
83433	2021	role birthplace gender displaydate prefix culture \ 0 Artist None male born 1969 None American displayname alphasort \ 0 Michael Cherney (also known as Qiu Mai 秋麥) Cherney, Michael name personid deathplace \ 0 Michael Cherney (also known as Qiu Mai 秋麥) 55097 None displayorder 0 1	male
83535	2020	role birthplace gender displaydate prefix culture displayname \ 0 Artist New York, NY female born 1951 None American Arlene Shechet alphasort name personid deathplace displayorder 0 Shechet, Arlene Arlene Shechet 65650 None 1	female
83854	2021	role birthplace gender displaydate prefix culture \ 0 Artist Warsaw, Poland male born 1943 None Polish displayname alphasort name personid \ 0 Krzysztof Wodiczko Wodiczko, Krzysztof Krzysztof Wodiczko 68777 deathplace displayorder 0 None 1	male
84444	2020	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
84445	2021	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
84481	2021	role birthplace gender displaydate prefix culture \ 0 Artist Vienna, Austria unknown born 1974 None Austrian displayname alphasort name personid deathplace \ 0 Svenja Deininger Deininger, Svenja Svenja Deininger 69073 None displayorder 0 1	unknown
85354	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Philadelphia, PA male born 1941 None American John Dowell alphasort name personid deathplace displayorder 0 Dowell, John John Dowell 18562 None 1	male
85355	2021	role birthplace gender displaydate prefix \ 0 Artist Colorado Springs, Colorado male born 1972 None culture displayname alphasort name \ 0 Mississippi Choctaw Jeffrey Gibson Gibson, Jeffrey Jeffrey Gibson personid deathplace displayorder 0 69254 None 1	male
85413	2020	role birthplace gender displaydate prefix culture \ 0 Artist Cologne, Germany unknown born 1960 None German displayname alphasort name personid deathplace \ 0 David Vostell Vostell, David David Vostell 69488 None displayorder 0 1	unknown
85537	2021	role birthplace gender displaydate prefix culture \ 0 Artist Detroit, Michigan unknown born 1987 None American displayname alphasort name personid deathplace displayorder 0 Jarod Lew Lew, Jarod Jarod Lew 69279 None 1	unknown
85538	2021	role birthplace gender displaydate prefix culture \ 0 Artist Little Rock, Arkansas unknown born 1983 None American displayname alphasort name personid deathplace \ 0 LaToya M. Hobbs Hobbs, LaToya M. LaToya M. Hobbs 69280 None displayorder 0 1	unknown
85539	2021	role birthplace gender displaydate prefix culture \ 0 Artist Little Rock, Arkansas unknown born 1983 None American displayname alphasort name personid deathplace \ 0 LaToya M. Hobbs Hobbs, LaToya M. LaToya M. Hobbs 69280 None displayorder 0 1	unknown
85540	2021	role birthplace gender displaydate prefix culture \ 0 Artist Little Rock, Arkansas unknown born 1983 None American displayname alphasort name personid deathplace \ 0 LaToya M. Hobbs Hobbs, LaToya M. LaToya M. Hobbs 69280 None displayorder 0 1	unknown
85541	2021	role birthplace gender displaydate prefix culture \ 0 Artist Little Rock, Arkansas unknown born 1983 None American displayname alphasort name personid deathplace \ 0 LaToya M. Hobbs Hobbs, LaToya M. LaToya M. Hobbs 69280 None displayorder 0 1	unknown
85542	2021	role birthplace gender displaydate prefix culture \ 0 Artist Little Rock, Arkansas unknown born 1983 None American displayname alphasort name personid deathplace \ 0 LaToya M. Hobbs Hobbs, LaToya M. LaToya M. Hobbs 69280 None displayorder 0 1	unknown
85543	2021	role birthplace gender displaydate prefix culture \ 0 Artist Little Rock, Arkansas unknown born 1983 None American displayname alphasort name personid deathplace \ 0 LaToya M. Hobbs Hobbs, LaToya M. LaToya M. Hobbs 69280 None displayorder 0 1	unknown
86141	2021	role birthplace gender displaydate prefix culture \ 0 Artist \r\nAccra, Ghana unknown born 1996 None Ghanaian displayname alphasort name personid deathplace \ 0 Claudio Eshun Eshun, Claudio Claudio Eshun 69294 None displayorder 0 1	unknown
86632	2021	role birthplace gender displaydate prefix culture \ 0 Artist Bogotá, Colombia unknown born 1983 None Colombian displayname alphasort name personid deathplace displayorder 0 Iván Argote Argote, Iván Iván Argote 69469 None 1	unknown
86703	2020	role birthplace gender displaydate prefix culture \ 0 Artist Cape Town, South Africa female born 1968 None South African displayname alphasort name personid deathplace displayorder 0 Lisa Brice Brice, Lisa Lisa Brice 69470 None 1	female
86705	2020	role birthplace gender displaydate prefix culture \ 0 Artist San Bernardino, California male born 1982 None American displayname alphasort name personid \ 0 Paul Mpagi Sepuya Sepuya, Paul Mpagi Paul Mpagi Sepuya 69485 deathplace displayorder 0 None 1	male
87739	2021	role birthplace gender displaydate prefix culture \ 0 Artist None unknown born 1958, Taipei, Taiwan None None displayname alphasort name personid deathplace displayorder 0 Yu-Wen Wu 吳育雯 Wu, Yu-Wen Yu-Wen Wu 吳育雯 69710 None 1	unknown
87811	2022	role birthplace gender displaydate prefix culture displayname \ 0 Artist None unknown born 1945 None None Park Dae Sung 박대성 alphasort name personid deathplace displayorder 0 Park, Dae Sung Park Dae Sung 박대성 69940 None 1	unknown
87816	2020	role birthplace gender displaydate prefix culture \ 0 Artist Mexico City, Mexico unknown born 1970 None Mexican displayname alphasort name personid deathplace displayorder 0 Bosco Sodi Sodi, Bosco Bosco Sodi 65348 None 1	unknown
87817	2021	role birthplace gender displaydate prefix culture \ 0 Artist Mexico City, Mexico unknown born 1970 None Mexican displayname alphasort name personid deathplace displayorder 0 Bosco Sodi Sodi, Bosco Bosco Sodi 65348 None 1	unknown
88168	2023	role birthplace gender displaydate prefix culture \ 0 Artist Prague, Czechoslovakia female born 1944 None German displayname alphasort name \ 0 Katharina Sieverding Sieverding, Katharina Katharina Sieverding personid deathplace displayorder 0 19111 None 1	female
88172	2021	role birthplace gender displaydate prefix culture \ 0 Artist Kolkata, India unknown born 1983 None Indian 1 Artist Basildon, England unknown None None British displayname alphasort name personid deathplace \ 0 Poulomi Basu Basu, Poulomi Poulomi Basu 70143 None 1 CJ Clarke Clarke, CJ CJ Clarke 70292 None displayorder 0 1 1 2	unknown
88176	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Hai Duong, Vietnam unknown born 1987 None German Sung Tieu alphasort name personid deathplace displayorder 0 Tieu, Sung Sung Tieu 70146 None 1	unknown
88193	2022	role birthplace gender displaydate prefix culture displayname \ 0 Artist Chappaqua, NY unknown born 1975 None American Gillian Laub alphasort name personid deathplace displayorder 0 Laub, Gillian Gillian Laub 65778 None 1	unknown
88241	2020	role birthplace gender displaydate prefix culture \ 0 Artist Umlazi, South Africa female born 1972 None South African displayname alphasort name personid deathplace \ 0 Zanele Muholi Muholi, Zanele Zanele Muholi 70193 None displayorder 0 1	female
88339	2020	role birthplace gender displaydate prefix culture \ 0 Artist Riverside, California unknown born 1990 None American displayname alphasort name personid deathplace \ 0 Johnnie Chatman Chatman, Johnnie Johnnie Chatman 70323 None displayorder 0 1	unknown
88341	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist India unknown born 1995 None Indian Cheryl Mukherji alphasort name personid deathplace displayorder 0 Mukherji, Cheryl Cheryl Mukherji 70324 None 1	unknown
88342	2022	role birthplace gender displaydate prefix culture displayname \ 0 Artist India unknown born 1995 None Indian Cheryl Mukherji alphasort name personid deathplace displayorder 0 Mukherji, Cheryl Cheryl Mukherji 70324 None 1	unknown
88437	2022	role birthplace gender displaydate prefix culture displayname \ 0 Artist None unknown born 1992 None American Trent Bozeman alphasort name personid deathplace displayorder 0 Bozeman, Trent Trent Bozeman 70337 None 1	unknown
88447	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Canada unknown born 1979 None Canadian Philip Cheung alphasort name personid deathplace displayorder 0 Cheung, Philip Philip Cheung 70430 None 1	unknown
88448	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Canada unknown born 1979 None Canadian Philip Cheung alphasort name personid deathplace displayorder 0 Cheung, Philip Philip Cheung 70430 None 1	unknown
88449	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Canada unknown born 1979 None Canadian Philip Cheung alphasort name personid deathplace displayorder 0 Cheung, Philip Philip Cheung 70430 None 1	unknown
88450	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Canada unknown born 1979 None Canadian Philip Cheung alphasort name personid deathplace displayorder 0 Cheung, Philip Philip Cheung 70430 None 1	unknown
88451	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Canada unknown born 1979 None Canadian Philip Cheung alphasort name personid deathplace displayorder 0 Cheung, Philip Philip Cheung 70430 None 1	unknown
88461	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Canada unknown born 1979 None Canadian Philip Cheung alphasort name personid deathplace displayorder 0 Cheung, Philip Philip Cheung 70430 None 1	unknown
88462	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Canada unknown born 1979 None Canadian Philip Cheung alphasort name personid deathplace displayorder 0 Cheung, Philip Philip Cheung 70430 None 1	unknown
88463	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Canada unknown born 1979 None Canadian Philip Cheung alphasort name personid deathplace displayorder 0 Cheung, Philip Philip Cheung 70430 None 1	unknown
88464	2021	role birthplace gender displaydate prefix culture \ 0 Artist Warsaw, Poland male born 1943 None Polish displayname alphasort name personid \ 0 Krzysztof Wodiczko Wodiczko, Krzysztof Krzysztof Wodiczko 68777 deathplace displayorder 0 None 1	male
88465	2021	role birthplace gender displaydate prefix culture \ 0 Artist Warsaw, Poland male born 1943 None Polish displayname alphasort name personid \ 0 Krzysztof Wodiczko Wodiczko, Krzysztof Krzysztof Wodiczko 68777 deathplace displayorder 0 None 1	male
88466	2021	role birthplace gender displaydate prefix culture \ 0 Artist Warsaw, Poland male born 1943 None Polish displayname alphasort name personid \ 0 Krzysztof Wodiczko Wodiczko, Krzysztof Krzysztof Wodiczko 68777 deathplace displayorder 0 None 1	male
88477	2023	role birthplace gender displaydate prefix culture \ 0 Artist Hokkaido, Japan female born 1990 None Japanese displayname alphasort name personid \ 0 Kawamura Sayaka 川村紗耶佳 Kawamura, Sayaka Kawamura Sayaka 川村紗耶佳 70511 deathplace displayorder 0 None 1	female
88478	2023	role birthplace gender displaydate prefix culture \ 0 Artist Hokkaido, Japan female born 1990 None Japanese displayname alphasort name personid \ 0 Kawamura Sayaka 川村紗耶佳 Kawamura, Sayaka Kawamura Sayaka 川村紗耶佳 70511 deathplace displayorder 0 None 1	female
88479	2023	role birthplace gender displaydate prefix culture \ 0 Artist Hokkaido, Japan female born 1990 None Japanese displayname alphasort name personid \ 0 Kawamura Sayaka 川村紗耶佳 Kawamura, Sayaka Kawamura Sayaka 川村紗耶佳 70511 deathplace displayorder 0 None 1	female
88561	2022	role birthplace gender displaydate prefix culture \ 0 Artist Enugu, Nigeria unknown born 1974 None German displayname alphasort name personid deathplace \ 0 Ngozi Schommers Schommers, Ngozi Ngozi Schommers 70622 None displayorder 0 1	unknown
88587	2023	role birthplace gender displaydate prefix \ 0 Artist \r\nDang Jin, South Korea \r\n female born 1970 None culture displayname alphasort name \ 0 American Jane Yang-D'Haene Yang-D'Haene, Jane Jane Yang-D'Haene personid deathplace displayorder 0 70754 None 1	female
88588	2023	role birthplace gender displaydate prefix \ 0 Artist \r\nDang Jin, South Korea \r\n female born 1970 None culture displayname alphasort name \ 0 American Jane Yang-D'Haene Yang-D'Haene, Jane Jane Yang-D'Haene personid deathplace displayorder 0 70754 None 1	female
88684	2023	role birthplace gender displaydate prefix culture \ 0 Artist Little Rock, Arkansas unknown born 1983 None American displayname alphasort name personid deathplace \ 0 LaToya M. Hobbs Hobbs, LaToya M. LaToya M. Hobbs 69280 None displayorder 0 1	unknown
88685	2023	role birthplace gender displaydate prefix culture displayname \ 0 Artist Conroe, Texas unknown born 1972 None American Delita Martin alphasort name personid deathplace displayorder 0 Martin, Delita Delita Martin 70794 None 1	unknown
88692	2021	role birthplace gender displaydate prefix culture \ 0 Artist Los Angeles, California female born 1956 None American displayname alphasort name personid deathplace displayorder 0 Alison Saar Saar, Alison Alison Saar 18837 None 1	female
88693	2023	role birthplace gender displaydate prefix culture \ 0 Artist Los Angeles, California female born 1956 None American displayname alphasort name personid deathplace displayorder 0 Alison Saar Saar, Alison Alison Saar 18837 None 1	female
88695	2023	role birthplace gender displaydate prefix culture \ 0 Artist St. Paul, Minnesota unknown born 1993 None American displayname alphasort name personid deathplace displayorder 0 Chase Hall Hall, Chase Chase Hall 70835 None 1	unknown
88709	2022	role birthplace gender displaydate prefix culture displayname \ 0 Artist None unknown None None None Chloe Alexander alphasort name personid deathplace displayorder 0 Alexander, Chloe Chloe Alexander 70868 None 1	unknown
88739	2023	role birthplace gender displaydate prefix culture displayname \ 0 Artist None unknown None None None Felekşan Onar alphasort name personid deathplace displayorder 0 Onar, Felekşan Felekşan Onar 70878 None 1	unknown
88751	2022	role birthplace gender displaydate prefix culture \ 0 Artist East Berlin, Germany female born 1953 None German displayname alphasort name personid \ 0 Cornelia Schleime Schleime, Cornelia Cornelia Schleime 69455 deathplace displayorder 0 None 1	female
88752	2022	role birthplace gender displaydate prefix culture \ 0 Artist East Berlin, Germany female born 1953 None German displayname alphasort name personid \ 0 Cornelia Schleime Schleime, Cornelia Cornelia Schleime 69455 deathplace displayorder 0 None 1	female
88753	2023	role birthplace gender displaydate prefix culture \ 0 Artist East Berlin, Germany female born 1953 None German displayname alphasort name personid \ 0 Cornelia Schleime Schleime, Cornelia Cornelia Schleime 69455 deathplace displayorder 0 None 1	female
88754	2022	role birthplace gender displaydate prefix culture \ 0 Artist East Berlin, Germany female born 1953 None German displayname alphasort name personid \ 0 Cornelia Schleime Schleime, Cornelia Cornelia Schleime 69455 deathplace displayorder 0 None 1	female
88755	2022	role birthplace gender displaydate prefix culture \ 0 Artist East Berlin, Germany female born 1953 None German displayname alphasort name personid \ 0 Cornelia Schleime Schleime, Cornelia Cornelia Schleime 69455 deathplace displayorder 0 None 1	female
88756	2023	role birthplace gender displaydate prefix culture \ 0 Artist East Berlin, Germany female born 1953 None German displayname alphasort name personid \ 0 Cornelia Schleime Schleime, Cornelia Cornelia Schleime 69455 deathplace displayorder 0 None 1	female
88757	2023	role birthplace gender displaydate prefix culture \ 0 Artist East Berlin, Germany female born 1953 None German displayname alphasort name personid \ 0 Cornelia Schleime Schleime, Cornelia Cornelia Schleime 69455 deathplace displayorder 0 None 1	female
88784	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist None female born 1951 None American Marcy Hermansader alphasort name personid deathplace displayorder 0 Hermansader, Marcy Marcy Hermansader 70966 None 1	female
88799	2022	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88800	2022	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88801	2022	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88802	2022	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88803	2022	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88804	2022	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88805	2023	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88806	2023	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88807	2023	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88808	2023	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88809	2023	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88810	2023	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88811	2023	role birthplace gender displaydate prefix culture \ 0 Artist Abidjan, Ivory Coast unknown born 1974 None Ivorian displayname alphasort name personid deathplace \ 0 Joana Choumali Choumali, Joana Joana Choumali 69012 None displayorder 0 1	unknown
88813	2021	role birthplace gender displaydate prefix culture \ 0 Artist New York, New York unknown born 1990 None Dominican American displayname alphasort name personid deathplace \ 0 Joiri Minaya Minaya, Joiri Joiri Minaya 70991 None displayorder 0 1	unknown
88814	2021	role birthplace gender displaydate prefix culture \ 0 Artist New York, New York unknown born 1990 None Dominican American displayname alphasort name personid deathplace \ 0 Joiri Minaya Minaya, Joiri Joiri Minaya 70991 None displayorder 0 1	unknown
88815	2022	role birthplace gender displaydate prefix culture \ 0 Artist New York, New York unknown born 1990 None Dominican American displayname alphasort name personid deathplace \ 0 Joiri Minaya Minaya, Joiri Joiri Minaya 70991 None displayorder 0 1	unknown
88816	2022	role birthplace gender displaydate prefix culture \ 0 Artist Browning, Montana unknown born 1989 None Piikani displayname alphasort name personid deathplace \ 0 Terran Last Gun Last Gun, Terran Terran Last Gun 70993 None displayorder 0 1	unknown
88817	2022	role birthplace gender displaydate prefix culture \ 0 Artist Browning, Montana unknown born 1989 None Piikani displayname alphasort name personid deathplace \ 0 Terran Last Gun Last Gun, Terran Terran Last Gun 70993 None displayorder 0 1	unknown
88820	2023	role birthplace gender displaydate prefix culture \ 0 Artist None unknown lives and works in Frankfurt None German displayname alphasort name personid deathplace displayorder 0 Aerin Hong Hong, Aerin Aerin Hong 71002 None 1	unknown
88821	2022	role birthplace gender displaydate prefix culture displayname \ 0 Artist Hai Duong, Vietnam unknown born 1987 None German Sung Tieu alphasort name personid deathplace displayorder 0 Tieu, Sung Sung Tieu 70146 None 1	unknown
88822	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Cologne, Germany male born 1969 None German Kota Ezawa alphasort name personid deathplace displayorder 0 Ezawa, Kota Kota Ezawa 71004 None 1	male
88823	2021	role birthplace gender displaydate prefix culture displayname \ 0 Artist Cologne, Germany male born 1969 None German Kota Ezawa alphasort name personid deathplace displayorder 0 Ezawa, Kota Kota Ezawa 71004 None 1	male
88856	2020	role birthplace gender displaydate prefix culture \ 0 Artist Toronto, Canada unknown born 1976 None Canadian displayname alphasort name personid deathplace \ 0 Saira McLaren McLaren, Saira Saira McLaren 71039 None displayorder 0 1	unknown
88857	2022	role birthplace gender displaydate prefix culture displayname \ 0 Artist Tehran, Iran unknown born 1985 None Iranian Nazanin Noroozi alphasort name personid deathplace displayorder 0 Noroozi, Nazanin Nazanin Noroozi 71040 None 1	unknown
88858	2021	role birthplace gender displaydate prefix culture \ 0 Artist Detroit, Michigan unknown born 1982 None American displayname alphasort name personid deathplace \ 0 LaKela Brown Brown, LaKela LaKela Brown 71041 None displayorder 0 1	unknown
88859	2023	role birthplace gender displaydate prefix culture \ 0 Artist Hollywood, Florida male born 1975 None American displayname alphasort name personid \ 0 William Villalongo Villalongo, William William Villalongo 71043 deathplace displayorder 0 None 1	male
88879	2023	role birthplace gender displaydate prefix culture \ 0 Artist Chicago, Illinois male born 1977 None American displayname alphasort name personid deathplace \ 0 Rashid Johnson Johnson, Rashid Rashid Johnson 68764 None displayorder 0 1	male
88997	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
88998	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
88999	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89000	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89001	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89002	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89003	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89004	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89005	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89006	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89007	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89008	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89009	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89010	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89011	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89012	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89013	2022	role birthplace gender displaydate prefix culture \ 0 Artist Lima, Peru unknown born 1972 None Peruvian displayname alphasort name \ 0 Sandra Gamarra Heshiki Gamarra Heshiki, Sandra Sandra Gamarra Heshiki personid deathplace displayorder 0 71247 None 1	unknown
89014	2022	role birthplace gender displaydate prefix culture \ 0 Artist São Paulo, Brazil unknown born 1982 None Tucano displayname alphasort name personid deathplace \ 0 Daiara Tukano Tukano, Daiara Daiara Tukano 71251 None displayorder 0 1	unknown
89015	2022	role birthplace gender displaydate prefix culture \ 0 Artist São Paulo, Brazil unknown born 1982 None Tucano displayname alphasort name personid deathplace \ 0 Daiara Tukano Tukano, Daiara Daiara Tukano 71251 None displayorder 0 1	unknown
89025	2022	role birthplace gender displaydate prefix culture \ 0 Artist São Paulo, Brazil female born 1967 None Brazilian displayname alphasort name personid deathplace \ 0 Rosana Paulino Paulino, Rosana Rosana Paulino 68809 None displayorder 0 1	female
89113	2023	role birthplace gender displaydate prefix culture \ 0 Artist Cutral-Có,\nArgentina unknown born 1981 None Argentinian displayname alphasort name personid deathplace \ 0 Jimena Croceri Croceri, Jimena Jimena Croceri 71339 None displayorder 0 1	unknown
89114	2022	role birthplace gender displaydate prefix culture \ 0 Artist Cutral-Có,\nArgentina unknown born 1981 None Argentinian displayname alphasort name personid deathplace \ 0 Jimena Croceri Croceri, Jimena Jimena Croceri 71339 None displayorder 0 1	unknown
89147	2023	role birthplace gender displaydate prefix culture \ 0 Artist Cutral-Có,\nArgentina unknown born 1981 None Argentinian displayname alphasort name personid deathplace \ 0 Jimena Croceri Croceri, Jimena Jimena Croceri 71339 None displayorder 0 1	unknown
89613	2023	role birthplace gender displaydate prefix culture \ 0 Artist Queens, New Yorks unknown born 1974 None American displayname alphasort name personid \ 0 Christopher Myers Myers, Christopher Christopher Myers 71451 deathplace displayorder 0 None 1	unknown
89631	2020	role birthplace gender displaydate prefix culture displayname \ 0 Artist Brooklyn, NY male born 1940 None American Arthur Tress alphasort name personid deathplace displayorder 0 Tress, Arthur Arthur Tress 71537 None 1	male
89633	2020	role birthplace gender displaydate prefix culture displayname \ 0 Artist Brooklyn, NY male born 1940 None American Arthur Tress alphasort name personid deathplace displayorder 0 Tress, Arthur Arthur Tress 71537 None 1	male
89634	2020	role birthplace gender displaydate prefix culture displayname \ 0 Artist Brooklyn, NY male born 1940 None American Arthur Tress alphasort name personid deathplace displayorder 0 Tress, Arthur Arthur Tress 71537 None 1	male
89635	2020	role birthplace gender displaydate prefix culture displayname \ 0 Artist Brooklyn, NY male born 1940 None American Arthur Tress alphasort name personid deathplace displayorder 0 Tress, Arthur Arthur Tress 71537 None 1	male
89666	2024	role birthplace gender displaydate prefix culture displayname \ 0 Artist Torrance,\nCA female born 1976 None American Raina Lee alphasort name personid deathplace displayorder 0 Raina Lee Raina Lee 71648 None 1	female
89667	2024	role birthplace gender displaydate prefix culture displayname \ 0 Artist Torrance,\nCA female born 1976 None American Raina Lee alphasort name personid deathplace displayorder 0 Raina Lee Raina Lee 71648 None 1	female
89668	2022	role birthplace gender displaydate prefix culture displayname \ 0 Artist Torrance,\nCA female born 1976 None American Raina Lee alphasort name personid deathplace displayorder 0 Raina Lee Raina Lee 71648 None 1	female
89669	2022	role birthplace gender displaydate prefix culture displayname \ 0 Artist Torrance,\nCA female born 1976 None American Raina Lee alphasort name personid deathplace displayorder 0 Raina Lee Raina Lee 71648 None 1	female
89670	2024	role birthplace gender displaydate prefix culture displayname \ 0 Artist Torrance,\nCA female born 1976 None American Raina Lee alphasort name personid deathplace displayorder 0 Raina Lee Raina Lee 71648 None 1	female
89671	2024	role birthplace gender displaydate prefix \ 0 Artist Shenyang,\nLiaoning province, China female born 1998 None culture displayname alphasort name personid \ 0 Chinese Danyang Song 宋丹楊 Song, Danyang Danyang Song 宋丹楊 71649 deathplace displayorder 0 None 1	female
89672	2024	role birthplace gender displaydate prefix \ 0 Artist Shenyang,\nLiaoning province, China female born 1998 None culture displayname alphasort name personid \ 0 Chinese Danyang Song 宋丹楊 Song, Danyang Danyang Song 宋丹楊 71649 deathplace displayorder 0 None 1	female
89673	2021	role birthplace gender displaydate prefix \ 0 Artist Shenyang,\nLiaoning province, China female born 1998 None culture displayname alphasort name personid \ 0 Chinese Danyang Song 宋丹楊 Song, Danyang Danyang Song 宋丹楊 71649 deathplace displayorder 0 None 1	female
90345	2023	None	unknown
90399	2023	role birthplace gender displaydate prefix culture \ 0 Artist Hokkaido, Japan female born 1990 None Japanese displayname alphasort name personid \ 0 Kawamura Sayaka 川村紗耶佳 Kawamura, Sayaka Kawamura Sayaka 川村紗耶佳 70511 deathplace displayorder 0 None 1	female
172458	2650	None	unknown

	title	year
18	From Mind, Heart, and Hand: Persian, Turkish, and Indian Drawings from SCWelch	2005.0
46	The Ottomans: Images from East and West	1989.0
218	Arabesque	1990.0
432	Eyes to the East: Indian, Persian, and Turkish Art Given by Harvard Graduates	1990.0
443	Five Masters of Persian and Indian Painting	1992.0
480	From India's Hills and Plains: Rajput Painting from the Punjab and Rajasthan, 17th through 19th Centuries	1993.0
496	Gods, Thrones, and Peacocks - Revisited: Northern Indian Miniatures from two Traditions, Fifteenth to Nineteenth Centuries	1990.0
521	Hot as Curry - Subtle as Moonlight: Masterpieces of Rajput Painting	1991.0
601	Linear Graces ... and Disgraces: Part I, Drawings from the Courts of Persia, Turkey, and India, 15th-19th Centuries	1994.0
602	Linear Graces ... and Disgraces: Part II, Drawings from the Courts of Persia, Turkey, and India, 15th-19th Centuries	1994.0
755	Paintings for Princes: The Art of the Book in Islam	1990.0
788	Poignant, Picturesque, and Berserk Northern Indian Paintings and Objects of the Seventeenth through Nineteenth Centuries	1992.0
803	Precisely to the Point, Daggers and Drawings from Persia and India: 15th-19th century	1992.0
952	The Grace of White: Indian Paintings and Objects, 16th-19th Centuries	1994.0
981	The Ottomans and their Contemporaries	1992.0
1255	Woven, Hammered, and Thrown: Textiles and Objects from the Islamic World	1991.0
1867	The Art of Kotah at Harvard	1997.0
1869	Arts of Empire: Mughal India, 1526-1705	1992.0
1870	Infancy to Old Age: A Gallery of Indians	1993.0
1872	Indian Harvest: Part 1	1995.0
1873	Indian Harvest: Part 2	1996.0
1878	The Edwin Binney 3rd Collection of Turkish Art at the Harvard University Art Museums	1987.0
2573	In Detail: Looking at Persian, Turkish and Indian Pictures	1995.0
3690	India: From Tribe to Court	1981.0
3691	From the Way of the Favored: Objects from the World of Islam	NaN

	title	year
19	HAA 1 Survey Course: Landmarks of World Art and Architecture [Spring 2007]	2007.0
90	32Q: 1520 Art in Germany Between the Wars (Interwar and Bauhaus)	2014.0
3093	Paintings by Max Beckmann from the Pinakothek der Moderne, Munich	2007.0
3169	Lyonel Feininger: Photographs, 1928-1939	2011.0
3382	Recent Acquisitions, Part I: Contemporary Photographs	2012.0
3730	European and American Pop Art, 1955-1975	2015.0
3745	A History of Photography	2015.0
3867	Modern Art and Modernity	2016.0
4059	What about Revolution? Aesthetic Practices after 1917	2017.0
4101	Looking Back: The Western Tradition in Retrospect	2018.0
4129	Adam and Eve	2018.0
4169	Hans Arp’s Constellations II	2019.0
4218	Displaying Latin America	2019.0
4280	Early Christian Africa: Arts of Transformation	2019.0
4281	Critical Printing	2019.0
4421	A Colloquium in the Visual Arts	2021.0
4450	Himalayan Art: Art of the Divine Abode	2022.0
4469	A Colloquium in the Visual Arts	2022.0
4492	Artisanal Modernism	2023.0
4509	A Colloquium in the Visual Arts	2023.0
4560	Picasso: War, Combat, and Revolution	2024.0
4578	A Colloquium in the Visual Arts	2024.0
4579	Art of the Black World	2024.0
4602	The Bauhaus and Harvard	2019.0
4973	The Art of Looking: 150 Years of Art History at Harvard	2025.0

	title	year
32	Paragons of Wisdom and Virtue: East Asian Figure Painting	1997.0
33	Rocks, Mountains, Landscapes, and Gardens: The Essence of East Asian Painting (1997)	1997.0
34	Art of the Fan: China, Korea, Japan	1995.0
35	Asian Art: Recent Acquisitions	1995.0
36	Women and the Arts of Asia	1994.0
37	East Asia in the Nineteenth Century	1994.0
38	The Arts of Korea	1992.0
39	Chinese and Japanese Painting of the Early Modern Era	1993.0
40	Later Chinese and Japanese Figure Painting in Decorative Arts	1992.0
41	Chinese Painting and Decorative Arts from the Permanent Collection	1991.0
3022	Cultivating Virtue: Botanical Motifs and Symbols in East Asian Art	2006.0
3077	Downtime	2007.0
3094	A Tradition Redefined: Modern and Contemporary Chinese Ink Paintings from the Chu-tsing Li Collection, 1950-2000	2007.0
3102	The New Chinese Landscape: Recent Acquisitions	2006.0
3149	Re-View: S228-230 Arts of Asia	2008.0
3192	Re-View: S228-230 (Asian rotation: 1)	2008.0
3202	Re-View: S228-230 (Asian rotation: 2)	2009.0
3214	Re-View: S228-230 (Asian rotation: 3)	2009.0
3215	Re-View: S228-230 (Asian rotation: 4)	2010.0
3258	Re-View: S228-230 (Asian rotation: 5) Brush and Ink Reconsidered: Contemporary Chinese Landscapes	2010.0
3327	Re-View: S228-230 (Asian rotation: 6)	2011.0
3383	Re-View: S228-230 (Asian rotation: 7) Art of the Fan: China, Korea, Japan	2011.0
3467	Cultivating Virtue: Botanical Motifs and Symbols in East Asian Art	2012.0

	title	year
96	32Q: 2550 Islamic	2014.0
97	32Q: 2590 South and Southeast Asia	2014.0
2568	The Enlightened Eye: Gifts from John Goelet	2000.0
2572	Courts and Countryside: Islamic Paintings from the 14th through the 17th Century	1999.0
2638	The Continuous Stroke of a Breath: Calligraphy from the Islamic World	2003.0
2858	The Sport of Kings: Art of the Hunt in Iran and India	2005.0
2874	The Tablet and the Pen: Drawings from the Islamic World	2006.0
2916	Closely Focused, Intensely Felt: Selections from the Norma Jean Calderwood Collection of Islamic Art	2004.0
3044	Overlapping Realms: Arts of the Islamic World and India, 900-1900	2006.0
3118	On the Path of Madness: Representations of Majnun in Persian, Turkish, and Indian Painting	2007.0
3148	Re-View: Arts of India & the Islamic Lands	2008.0
3193	Re-View: S231 (Islamic rotation) Tree of Life	2008.0
3194	Re-View: S231 (Islamic rotation: 1) Color in the Oasis	2008.0
3252	Re-View: S231 (Islamic rotation: 4) Strolling Through Isfahan: Seventeenth-Century Paintings From Safavid Iran	2010.0
3255	Re-View: S231 (Islamic rotation: 5) Heroic Gestes: Epic Tales from Firdawsi's Shahnama	2010.0
3289	Re-View: S231 (Islamic rotation: 6) I was not Waving but Drowning	2010.0
3322	Re-View: S231 (Islamic rotation: 7) Company to Crown	2011.0
3542	African Art: The Art of Assemblage	2014.0
3783	Technologies of the Image: Art in 19th-Century Iran	2017.0
4222	Social Fabrics: Inscribed Textiles from Medieval Egyptian Tombs	2022.0
4299	A Decade of Collecting: Recent Acquisitions of Islamic and Later Indian Art	2000.0
4594	In Harmony: The Norma Jean Calderwood Collection of Islamic Art	2013.0

	title	year
21	Re-View: S422-423 Western Art of the Middle Ages & Renaissance	2008.0
63	32Q: 1220 Wertheim	2014.0
65	32Q: 2710 North Arcade	2014.0
66	32Q: 2700 Impressionism	2014.0
67	32Q: 2460 East Arcade	2014.0
68	32Q: 2440 Medieval	2014.0
69	32Q: 2410 South Arcade	2014.0
70	32Q: 2400 French/Italian/Spanish	2014.0
71	32Q: 2340 Cabinet Gallery	2014.0
72	32Q: 2300 Dutch & Flemish	2014.0
73	32Q: 2240 18th Century	2014.0
74	32Q: 2220 18th-19th Century	2014.0
75	32Q: 2210 West Arcade	2014.0
76	32Q: 2200 19th Century	2014.0
77	32Q: 2130 19th Century	2014.0
78	32Q: 2120 19th Century	2014.0
79	32Q: 2100 19th Century	2014.0
91	32Q: 2540 Renaissance	2014.0
92	32Q: 2500 Renaissance	2014.0
93	32Q: 2520 Winter Garden	2014.0
2639	A Private Passion: 19th-Century Paintings and Drawings from the Grenville L. Winthrop Collection, Harvard University	2003.0
3286	Re-View: S425A (Small Niche) The Art of Deceit: Looking at French Trompe L'Oeil	2010.0

	title	year
51	Re-View: S118 Paper Rotation 1 (Sept 2011)	2011.0
80	32Q: 1100	2014.0
82	32Q: 1120 Contemporary Art	2014.0
83	32Q: 1200 Gesture	2014.0
85	32Q: 1310 Surrealism	2014.0
3317	Re-View: European and American Art Since 1900	2011.0
3715	Everywhen: The Eternal Present in Indigenous Art from Australia	2016.0
3841	Doris Salcedo: The Materiality of Mourning	2016.0
4220	Crossing Lines, Constructing Home: Displacement and Belonging in Contemporary Art	2019.0
4430	Krzysztof Wodiczko: Portrait	2021.0
4494	Bosco Sodi: Origen	2023.0
4597	Mark Rothko's Harvard Murals	2014.0
4603	Fernando Bryce: The Book of Needs	2018.0
4604	Nam June Paik: Screen Play	2018.0

	title	year
26	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 08 Rotation 1)	2008.0
27	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 08 Rotation 3)	2008.0
3127	HAA 10 Survey Course: The Western Tradition, Art Since the Renaissance	2008.0
3203	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 08 Rotation 2)	2008.0
3216	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 09 Rotation 1)	2009.0
3217	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 09 Rotation 2)	2009.0
3218	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 09 Rotation 3)	2009.0
3257	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 10 Rotation 1)	2010.0
3300	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 10 Rotation 2)	2010.0
3301	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 10 Rotation 3)	2010.0
3379	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 11 Rotation 1)	2011.0
3380	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 11 Rotation 2)	2011.0
3381	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 11 Rotation 3)	2011.0

	title	year
53	32Q: 3740 Egyptian	2014.0
54	32Q: 3460 East Arcade	2014.0
55	32Q: 3400 Greek	2014.0
56	32Q: 3710 North Arcade	2014.0
57	32Q: 3200 West Arcade	2014.0
58	32Q: 3410 South Arcade	2014.0
59	32Q: 3700 Roman	2014.0
60	32Q: 3440 Middle East	2014.0
3071	Gods in Color: Painted Sculpture of Classical Antiquity	2007.0
3147	Re-View: S422 Ancient & Byzantine Art & Numismatics	2008.0
3447	HAA132e The Ideal of the Everyday in Greek Art (S427) Spring 2012	2012.0
3897	Animal-Shaped Vessels from the Ancient World: Feasting with Gods, Heroes, and Kings	2018.0
4442	Funerary Portraits from Roman Egypt: Facing Forward	2022.0

	title	year
64	32Q: 1440 Secessionism: Munich, Vienna, Berlin (Expressionism)	2014.0
88	32Q: 1500 Art in Germany Between the Wars (Expressionism-Interwar)	2014.0
89	32Q: 1510 German Expressionist Sculpture	2014.0
3708	Rebecca Horn: “Work in Progress”	2014.0
3842	Private Practice	2015.0
3939	Inventur—Art in Germany, 1943–55	2018.0
4434	Disrupt the View: Arlene Shechet at the Harvard Art Museums	2022.0
4600	Folding, Refraction, Touch: Modern and Contemporary Art in Dialogue with Wolfgang Tillmans	2016.0
4607	White Shadows: Anneliese Hager and the Camera-less Photograph	2022.0
4970	Made in Germany? Art and Identity in a Global Nation	2024.0

	title	year
3254	Re-View: S426A (Large Niche) #1: The Past and the Present: British Art of the 19th Century	2010.0
3290	Re-View: S426A (Large Niche) #2: The Past and the Present: British Art of the 19th Century	2010.0
3320	HAA 176m Berlin and Moscow, 1918/1930 Rotation #1 (S421) Spring 2011	2011.0
3563	Strategies of Exhibition Design in the Twentieth Century: History and Theory	2013.0
3564	Word and Image in Persian Painting	2013.0
3565	The Art of Looking	2013.0
3566	Civil War Project	2013.0
3567	Boston Taste, 1670 - 1970	2013.0
4468	A World Within Reach: Greek and Roman Art from the Loeb Collection	2023.0

	title	year
61	32Q: 1740 Early China I	2014.0
62	32Q: 2740 Buddhist II	2014.0
94	32Q: 1600 Early China II	2014.0
95	32Q: 1610 Buddhist Art I	2014.0
98	32Q: 2620 Winter Garden	2014.0
99	32Q: 2600 East Asian, Japanese, Chinese and Korean	2014.0
100	32Q: 1006 East Arcade	2014.0
3937	Adorning the Inner Court: Jun Ware for the Chinese Palace	2017.0

	title	year
3195	Re-View: S231 (Islamic rotation: 2) Sacred Sites: Nineteenth Century Photographs of Jerusalem	2009.0
3251	Re-View: S426A (Large Niche) HAA175w Pop Art	2010.0
3253	Around Antique: Prints, Drawings, and Photographs (Teaching Gallery) S421	2010.0
3463	Laurel Nakadate: Say You Love Me	2011.0
3466	Recent Acquisitions, Part II: Building the Collection	2012.0
4596	Recent Acquisitions, Part III: Kerry James Marshall	2012.0

	title	year
3781	Drawing: The Invention of a Modern Medium	2017.0
4311	Dare to Know: Prints and Drawings in the Age of Enlightenment	2022.0
4510	LaToya M. Hobbs: It’s Time	2024.0
4605	States of Play: Prints from Rembrandt to Delsarte	2021.0
4608	Prints from the Brandywine Workshop and Archives: Creative Communities	2022.0
4972	Edvard Munch: Technically Speaking	2025.0

	title	year
81	32Q: 1110	2014.0
86	32Q: 1320 Social Realism	2014.0
87	32Q: 1330	2014.0
3054	Classified Documents: The Social Museum of Harvard University, 1903-1931	2007.0
3738	Jesse Aron Green: Ärztliche Zimmergymnastik	2015.0

	title	year
3034	DISSENT!	2006.0
3101	Kara Walker: "Harper's Pictorial History of the Civil War (Annotated)"	2007.0
4593	Prints and the Pursuit of Knowledge in Early Modern Europe	2011.0
4595	Jasper Johns / In Press: The Crosshatch Works and the Logic of Print	2012.0
4598	Corita Kent and the Language of Pop	2015.0

	title	year
84	32Q: 1300 Early Modernism	2014.0
3150	Re-View: S427 Impressionist & Postimpressionist Art	2008.0
3717	The Dorothy and Herbert Vogel Collection: Fifty Works for Fifty States	2015.0
4601	Reverie: Christopher Wilmarth, Before and After Mallarmé	2017.0

	title	year
3030	"A Public Patriotic Museum": Artworks and Artifacts from the General Artemas Ward House	2006.0
3313	Tangible Things	2011.0
4388	Investigating the Renaissance	1996.0

	title	year
48	Re-View: S425A (Small Niche) Peter Blume	2010.0
3029	Under Cover: Artists' Sketchbooks	2006.0
3741	Body Image in French Art and Visual Culture (18th and 19th Centuries)	2016.0
3758	Drawings from the Age of Bruegel, Rubens, and Rembrandt	2016.0
4073	The Art of Drawing in the Early Dutch Golden Age, 1590–1630: Selected Works from the Abrams Collection	2017.0

	title	year
1832	Rasika, the Discerning Connoisseur: Indian Paintings from the John Kenneth Galbraith Collection	1998.0
1866	Shadows of God On Earth: Arts of the Ottoman, Safavid, and Mughal Dynasties	1997.0
1933	Divinely Inspired: Images of Mystics and Mendicants	1999.0
2331	The Sensuous and the Sublime: Representations of Love in the Arts of the Middle East and Southern Asia	2001.0
2571	A Grand Legacy: Arts of the Ottoman Empire	1999.0

	title	year
47	Re-View: S425A (Small Niche) New Acquisition: Lorna Simpson	2010.0
3092	Contemporary Art from the Harvard University Art Museums Collections	2007.0
3112	Long Life Cool White: Photographs by Moyra Davey	2008.0
3220	ACT UP New York: Activism, Art and the AIDS Crisis, 1987-1993	2009.0

	title	year
22	Re-View: S426A (Large Niche) 19th Century French Portrait Drawings	2009.0
2494	Bruegel to Rembrandt: Dutch and Flemish Drawings from the Maida and George Abrams Collection	2002.0
3032	Rembrandt and the Aesthetics of Technique	2006.0
3181	The Draughtsman at Work. Drawing in the Golden Century of Dutch Art	1980.0

	title	year
3105	Light Display Machines: Two Works by László Moholy-Nagy	2007.0
3153	Re-View: S118 European & American Art since 1900	2008.0
3168	Lyonel Feininger: Drawings and Watercolors from the William S. Lieberman Bequest to the Busch-Reisinger Museum	2011.0
4254	El Lissitzky 1890-1941	1987.0

	title	year
31	Re-View: S426A (Large Niche) #2: CB30 Photography and Society & HAA172w American Art and Modernity	2009.0
52	GenEd US12 American Encounters: Art, Contact, and Conflict, 1560-1860 (S427) Spring 2012	2012.0
3519	American Art and Modernity, 1865–1965	2012.0
3562	Modern Art and Modernity	2013.0

	title	year
28	Re-View: S426A (Large Niche) HAA Lit & Arts B24	2008.0
30	Re-View: S426A (Large Niche) #1: CB30 Photography and Society & HAA172w American Art and Modernity	2009.0
3446	HAA178m Cold War Photography (S427) Spring 2012	2012.0
3518	Seeing Is Believing: A History of Photography	2012.0

	title	year
309	Colors of the Caucasus	1990.0
363	Earthly Paradise: Gardens in Islamic Art	1993.0
768	Pattern and Purpose. Decorative Arts of Islam.	1994.0

	title	year
103	Recent Acquisitions 2020	2021.0
4462	Earthly Delights: 6,000 Years of Asian Ceramics	2022.0
4557	Future Minded: New Works in the Collection	2024.0

	title	year
3490	Flowing Streams and Rivers in Japanese Life	2006.0
3495	Great Waves: Chinese Themes in the Arts of Korea and Japan	2003.0
3514	Through the Seasons: Japanese Art in Nature	1999.0

Harvard Art Museum Data Analysis Project¶

People¶

Objective¶

In the inital analysis we start by identifying some basic properties of the art object records:¶

Below we can see when objects were made and acquired.¶

There are 246,377 objects in the HAM database as of July 5, 2025.¶

We see that 30% of the records have a 0 (null) creation date.¶

And that 29% have a null value for accession year.¶

70738 (30%) objects have a 0 value created date. 74150 (29%) objects have no accession year¶

What divsions are objects from and what kind of objects are they?¶

What cultures are objects from?¶

Pulling records with offensive keywords.¶

One group of records that immediately comes up is the Social Museum Collection. An important record of American history... derogatory keywords such as: Negroes, Jews, Indians, Cripples, Insane, Destitute occur with some regularity in the titles of these images with no disclaimer.¶

Examples of Records that contain derogatory terms¶

Example of records where an attempt has been made to deal with derogatory terms¶

Identification and Creation

Physical Descriptions

Acquisition and Rights

Publication History

Exhibition History

Subjects and Contexts

Related Digital Tours

The Social Museum Collection

Verification Level

Identification and Creation

Physical Descriptions

Acquisition and Rights

Exhibition History

Verification Level

Identification and Creation

Physical Descriptions

Acquisition and Rights

Verification Level

We also looked at representation within the Art collections. We started by looking at the Gender of Artists.¶

After some analysis, we see that of the approximately 250K items, 2.3% are female-attributed, 61.4% are male-attributed,and the remaining 35.97% are attributed to an unknown gender.¶

Below we plot artist's gender by year. Some observations are:¶

Looking a little more closely at gender trends in the last 75 years we see:¶

A dramatic Increase in proportion of female-attributed items¶

Why was there a sudden drop in the percent of number of artworks made by men in the 2020s?¶

Let's do the same alaysis but with objects that are eventually exhibited.¶

We see the same shift here but much more gradual.¶

Looking at what 'culture' female artists are from¶

We can look at trends in the gender of artists exhibited each year¶

We also looked at trends in Exhibitions and Acquisitions¶

Looking at exhibition curators¶

We see 93% of exhibitions do not have a curator listed.¶

Now we can see each major curator alongside of the tiles and years of the shows that they put on.¶

Trends in culture and divisions of objects acquired.¶

Now we will try to group exhibitions by descriptions and titles to better understand trends¶

Creating Word Clouds¶

Clustering Exhibition Descriptions¶

Try the same thing for titles¶

Lets look at the most exhibited artists¶

Now we use AWS Face recognition to talk about the people represented in the art objects.¶

We have data on 286,356 faces found in the collection¶

Recommendations:¶

	title	year
3208	Re-View: S426A (Large Niche) HAA171x Exoticism and Orientalism	2009.0
3318	HAA 271x The Origins of Modernity: The "New" 18th Century Rotation #1: Large Niche (S426A) Spring 2011	2011.0
3319	HAA 271x The Origins of Modernity: The "New" 18th Century Rotation #2: Large Niche (S426A) Spring 2011	2011.0

	title	year
42	Sacred Symbols of Islam	1980.0
43	The Hofer Bequest and Islamic Decorative Arts from the Fogg Collection	1980.0
44	Architectural Themes in the Arts of Islam	1981.0

	title	year
3520	Landmarks of World Art and Architecture	2012.0
3623	World’s Fairs	2014.0
3747	Art of Jazz: Form	2016.0

	title	year
4170	Prince Shōtoku: The Secrets Within	2019.0
4197	Japan on Paper	2019.0
4606	Painting Edo: Japanese Art from the Feinberg Collection	2020.0

	name	year	title
2375	None	1948	Seventy Master Drawings: A Loan Exhibition arranged in honor of Professor Paul J. Sachs on the occasion of his seventieth birthday
3379	Henri Zerner	2011	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 11 Rotation 1)
3380	Henri Zerner	2011	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 11 Rotation 2)
3381	Henri Zerner	2011	HAA 10 Survey Course (S421): The Western Tradition: Art Since the Renaissance (Fall 11 Rotation 3)
3446	Robin Kelsey	2012	HAA178m Cold War Photography (S427) Spring 2012
3447	Susanne Ebbinghaus	2012	HAA132e The Ideal of the Everyday in Greek Art (S427) Spring 2012

	name	year	title
3318	Ewa Lajer-Burcharth	2011	HAA 271x The Origins of Modernity: The "New" 18th Century Rotation #1: Large Niche (S426A) Spring 2011
3319	Ewa Lajer-Burcharth	2011	HAA 271x The Origins of Modernity: The "New" 18th Century Rotation #2: Large Niche (S426A) Spring 2011
3320	Amy Brauer	2011	HAA 176m Berlin and Moscow, 1918/1930 Rotation #1 (S421) Spring 2011
3518	Robin Kelsey	2012	Seeing Is Believing: A History of Photography
3519	Jennifer Roberts	2012	American Art and Modernity, 1865–1965
3520	Suzanne Blier	2012	Landmarks of World Art and Architecture
3561	Eugene Wang	2013	Voices in Chinese Painting
3562	Jennifer Roberts	2013	Modern Art and Modernity
3563	Amy Brauer	2013	Strategies of Exhibition Design in the Twentieth Century: History and Theory
3564	Amy Brauer	2013	Word and Image in Persian Painting
3565	Amy Brauer	2013	The Art of Looking

	name	year	title
3289	Mary McWilliams	2010	Re-View: S231 (Islamic rotation: 6) I was not Waving but Drowning
3783	Mary McWilliams	2017	Technologies of the Image: Art in 19th-Century Iran
4222	Mary McWilliams	2022	Social Fabrics: Inscribed Textiles from Medieval Egyptian Tombs
4594	Mary McWilliams	2013	In Harmony: The Norma Jean Calderwood Collection of Islamic Art

	name	year	title
3071	Susanne Ebbinghaus	2007	Gods in Color: Painted Sculpture of Classical Antiquity
3321	Carmen Arnold-Biucchi	2011	Re-View: S422 (Ancient rotation) "The Scholar as Collector: Margarete Bieber (1879-1978)"
3462	None	2006	Hellenistic Art: Objects from an Expanded World
3897	Susanne Ebbinghaus	2018	Animal-Shaped Vessels from the Ancient World: Feasting with Gods, Heroes, and Kings
4442	Susanne Ebbinghaus	2022	Funerary Portraits from Roman Egypt: Facing Forward
4468	Amy Brauer	2023	A World Within Reach: Greek and Roman Art from the Loeb Collection
4487	Jennifer Thum	2023	Seeing in Art and Medicine
4597	Mary Schneider Enriquez	2014	Mark Rothko's Harvard Murals
4972	Elizabeth M. Rudy	2025	Edvard Munch: Technically Speaking

	name	year	title
1867	Stuart Cary Welch	1997	The Art of Kotah at Harvard
2518	None	2003	Image and Empire: Picturing India during the Colonial Era
2559	Alexander Keefe	2003	Where Traditions Meet: Painting in India from the 15th-17th Centuries
2930	Jeffrey Spurr	2005	Silver and Shawls: India, Europe, and the Colonial Art Market
3193	Mary McWilliams	2008	Re-View: S231 (Islamic rotation) Tree of Life
3361	Katherine Eremin	2011	Beyond the Surface: Scientific Approaches to Islamic Metalwork

	name	year	title
2498	None	2002	Modern Quotidian: Furniture by Prouve, Perriand, Le Corbusier, Rietveld
2742	None	2004	Design~Recline: Modern Architecture and the Mid-Century Chaise Longue
3180	None	2008	Renzo Piano's Art Museum for Harvard
3464	Sanford Kwinter	2011	The Divine Comedy
4218	Laura Muir	2019	Displaying Latin America
4430	Mary Schneider Enriquez	2021	Krzysztof Wodiczko: Portrait
4974	Mitra Abbaspour	2025	Joana Choumali: Languages of West African Marketplaces